Audio Science Review of Expert 200.

[ICUDoc]

I don't give a damn about power output at that frequency.
I can't hear it, and what I do hear sounds great

[whatmore]

I don't give a damn about power output at that frequency.
I can't hear it, and what I do hear sounds great

I give a big damn if I’ve been misled.

[thumb5]

... the amp does not have to be capable of delivering rated output for a sine wave signal at any frequency in order to meet its rated output spec with a broadband signal or with music.

While this is true for all practical purposes (listening to music) as you pointed out, apparently the reduced power output above 10 kHz violates the US FTC standard for home audio amplifier power specifications.  Amir points to clause 432.3(e) which states:

(e) Rated power shall be obtainable at all frequencies within the rated power band without exceeding the rated maximum percentage of total harmonic distortion after input signals at said frequencies have been continuously applied at full rated power for not less than five (5) minutes at the amplifier's auxiliary input, or if not provided, at the phono input.

That seems fairly clear, so at least in the US it could be argued that the measured behaviour of the 200 is a regulatory concern even if it is of no practical consequence.

Personally this doesn't bother me at all from a listening-to-music point of view, for the reasons you summarised, although I would like to understand the technical rationale for the behaviour that several reviewers have now noted.  My hunch/guess at the moment is that limited power delivery at high frequencies is an inherent "feature" of the ADH architecture and the amp is protecting itself rather than the loudspeakers.

[Gerronwithit]

I don't give a damn about power output at that frequency.
I can't hear it, and what I do hear sounds great

I give a big damn if I’ve been misled.

I hope to God you gave your equipment a listen before you bought it and only at that point decided that Devialet was for you. I have been running mine for almost four years with no issues. There might be, and probably are better sounding amps but only my ears will be allowed to discern that, not a review or an oscilloscope. Or did your own ears deceive you?

[whatmore]

I hope to God you gave your equipment a listen before you bought it and only at that point decided that Devialet was for you. I have been running mine for almost four years with no issues. There might be, and probably are better sounding amps but only my ears will be allowed to discern that, not a review or an oscilloscope. Or did your own ears deceive you?

As I mentioned before, my devialet went up for sale the same day I tried it at home - it sounded like doggie doo.
Then someone suggested I turn off DPM and that made a "night and day difference".

But… that is not my point….
I'm happy that people like their Devialets. Truly I am
But that is not my point either….

My point is that there appears to be a big difference between what Devialet advertise their gear as being able to do and what it can actually do.
I feel that they have to explain this to their customers - current and potential.
That's all my point is..

Again, I'm not questioning how anyone feels about the sound of their amp…my gripe is not about that (although I do wonder what is broken with DPM)

[Confused]

Looking at the Paul Miller measurements for the D170 (page 5 of this thread) I spotted something that I found quite intriguing.

For the USB input:
Signal-to-Noise Ratio (A-weighted, 3rd-Octave mode) into 8ohm = 97.3dB

For the SPDIF input:
Signal-to-Noise Ratio (A-weighted, 3rd-Octave mode) into 8ohm = 117.7dB

Reasonable measurements both, but I am intrigued by the20dB difference between the USB and SPDIF input.

Subjectively, a few of us have observed in the past that we prefer the sound via AES/EBI with the "pre-Pro" Expert. In my own case, I preferred the Aurender N10 via AES/EBU, rather than via USB, and I preferred my microRendu via the Mutec MC3+USB, AES/EBU out, rather than via USB direct. OK - many variables here, but it does make me wonder what the above difference in S/N ratio makes to sound quality. Is there a correlation? Or maybe it is the myriad of other factors that lead to the subjectively observed difference in sound quality? Possibly a bit of both?

I am also puzzled at the 20dB difference in the headline result number, when the associated figures in the table below look far more similar. Can anyone explain or interpret this? I am at the "a little knowledge is a dangerous thing" level, when it comes to interpreting the detail here.

[petrik]

I happened to notice interesting results regarding jitter on the Miller’s measurements (96 kHz/24-bit). These might not be fully comparable results but the difference there is quite significant.
D-Premier: 49 psec
D170: ~2300 psec

Moreover, the difference is almost nonexisting at 48 kHz/24-bit.

[Confused]

I happened to notice interesting results regarding jitter on the Miller’s measurements (96 kHz/24-bit). These might not be fully comparable results but the difference there is quite significant.
D-Premier: 49 psec
D170: ~2300 psec

Moreover, the difference is almost nonexisting at 48 kHz/24-bit.

Looking at the measurements, it would seem that you are correct.  I also note that the "rouge" 96kHz/24-bit Jitter measurement for the D170 is only for the S/PDIF input, there is no equivalent measurement for USB. 

There is some explanation re jitter in the side bar of the original Hifi News review, see link below.  There is also some commentary regarding distortion, the difference between the analogue and digital inputs, very interesting and well worth a read.

 https://www.audio-components.de/assets/U...i-News.pdf

I also managed to find this, the Hifi News review of the Original D'Altelier.  Again, some interesting comments re the difference between the analogue and digital inputs, jitter and distortion at very high frequencies.  Although it occurs to me that nobody has a truly original Original D'Altllier anymore, as I presume they have all been upgraded to 1000 Pro internals now?

https://www.sarte-audio.com/sites/defaul..._hires.pdf

In more practical terms, I have had a difficult week and the UK is on "Lockdown".  But today is a public holiday, the sun is shining, I plan to open a bottle of wine, play an extensive collection of hard core electronica, and see if I can find a strange tune that actually features a 20kHz sine wave as part of the music and blow my 1000 Pro to pieces. 

I'll let you all know how this goes. 

[Confused]

Given Experts design topology, how can any of these measurements separate DAC performance from amplifier performance?

This is an excellent point, definitively they cannot. 

Consider you are measuring a "conventional" integrated amplifier, here you can measure just the amplifier, but to get measurements directly comparable to a digitally fed Devialet, you would need to add a DAC, which presumably adds its own noise, distortion and whatever else.

Indeed, in the case of the ASR measurements, or other measurements made via the analogue input, you are inevitably adding any noise or distortions from the ADC, which most of us do not need to use.  Clearly this penalises the Devialet results if considered in direct comparison with a conventional amplifier.

I suppose you could make direct comparisons of the digital input with integrated amps like the Gryphon or Hegal, which include their own DACs, but essentially these are just amp and DAC in the same box, a little different to Devialets ADH where the DAC is integral to the amplifiers performance.

A further step beyond all this is that with the typical usage for many, the Devialet is also a source as AIR or the CI board is effectively a "networked attached end point".

I do find the measurements interesting though.  As an example, I know many Devialet users get excellent subjective results using an external phono stage, which is fair enough when you consider that the built in phono stage is audibly noisy with low output MC's.  However, seeing how the various inputs measure, an external digital phono stage with S/PDIF output might be even better.  In theory at least.  It might even be possible to use a conventional analogue phono stage with an external ADC, which might yield good results if the external ADC measures better than Devialet's internal offering.  Again, in theory at least.

As an aside, I'm using the "listening with ears" method this morning.  No issues detected so far …….  

[PeppaPig]

Given Experts design topology, how can any of these measurements separate DAC performance from amplifier performance?

This is an excellent point, definitively they cannot. 

Consider you are measuring a "conventional" integrated amplifier, here you can measure just the amplifier, but to get measurements directly comparable to a digitally fed Devialet, you would need to add a DAC, which presumably adds its own noise, distortion and whatever else.

Indeed, in the case of the ASR measurements, or other measurements made via the analogue input, you are inevitably adding any noise or distortions from the ADC, which most of us do not need to use.  Clearly this penalises the Devialet results if considered in direct comparison with a conventional amplifier.

I suppose you could make direct comparisons of the digital input with integrated amps like the Gryphon or Hegal, which include their own DACs, but essentially these are just amp and DAC in the same box, a little different to Devialets ADH where the DAC is integral to the amplifiers performance.

A further step beyond all this is that with the typical usage for many, the Devialet is also a source as AIR or the CI board is effectively a "networked attached end point".

I do find the measurements interesting though.  As an example, I know many Devialet users get excellent subjective results using an external phono stage, which is fair enough when you consider that the built in phono stage is audibly noisy with low output MC's.  However, seeing how the various inputs measure, an external digital phono stage with S/PDIF output might be even better.  In theory at least.  It might even be possible to use a conventional analogue phono stage with an external ADC, which might yield good results if the external ADC measures better than Devialet's internal offering.  Again, in theory at least.

As an aside, I'm using the "listening with ears" method this morning.  No issues detected so far …….  

Hi Confused,

I am not really getting why the measurement of analog input is noisy. I have a turntable connected to my 1000 Pro, no matter the MM cartridge I used before or the currentC cartridge. Both is very very quite, friend came to my house listen to it are surprised Vinly can be this clean. The Devialet background is like black black black, it’s one of the thing it surprised me. So I don’t know what those measurements reflected in real world listening.