WiiM Amp Pro Streaming Stereo Amplifier Review

[Tell]

I do not like this amplifier and would not give it my recommendation.

1) load dependency is unknown as it was measured only with resistive load
2) distortion rises with frequency too much
3) IMD 19+20kHz distortion is very poor, the odd order skirts are most likely a case of output inductors non-linearity
4) power cube behaviour is poor

Another toy instead of a serious amplifier.

But in the end it'll still proably sound just as good as any class AB amplifier.

 

[Petrushka]

There are mysterious legal reasons for the lack of airplay.

It could be something as simple as not designing the motherboard to accommodate the Apple chip. Or it could be the licensing fee, or it could be that Apple just isn’t going to license airplay broadcasting.

 

[Tovarich007]

Tu as raison. J'aurais dû l'écrire comme ça, mais bon,

You worte : "All in all, the Wiim Ultra is a pretty good streamer/dac/preamp, with good performances, a very reasonable price tag and superb esthetics.
Neverthesess, it lacks some useful features for a model called "Ultra" : no Airplay connectivity (a very surprising omission, the lower models including it), no balanced analog output and subwoofer output, a good but suboptimal heaphone section."

It is at least very ambiguous as a formulation for those who will not check and would not know that there is a sub output... He should have written : "neither L and R outputs nor balanced sub outputs".

Mais nous avons levé toute ambiguité sur le sujet.

j'aurais en effet dû l'écrire comme tu le suggères.
Mais tu sais, sur les forums j'écris en vitesse et c'est quand même moins simple pour moi que le faire dans la langue de Molière.
Sur ce, je vais aller me balader, y'a pas que la musique et la hifi dans la vie.

 

[Sokel]

But in the end it'll still proably sound just as good as any class AB amplifier.

With 11-bits of IMD that's debatable.
IMD is the the most audible kind of distortion,that needs searching.

 

[amper42]

There are mysterious legal reasons for the lack of airplay.

It could be something as simple as not designing the motherboard to accommodate the Apple chip. Or it could be the licensing fee, or it could be that Apple just isn’t going to license airplay broadcasting.

It's funny how it's OK for basic electronic services like Airplay broadcasting to be restricted while Gmail is allowed to sniff through every sentence in your private Gmail to overwhelm us with targeted advertising. There should be basic use rights for consumers that are superior to what we currently have. Maybe if Apple was allowed to sniff through your entire music collection to discover what to sell you next they would allow Airplay broadcasting?

 

[NTK]

With 11-bits of IMD that's debatable.
IMD is the the most audible kind of distortion,that needs searching.

Dr David Griesinger did a study on the audibility of intermodulation distortions. His study was for loudspeakers, but since he had amplifier in the chain, he also looked at its effects. Here are his conclusion: (http://www.davidgriesinger.com/intermod.ppt)

And what he meant by "high quality amplifier".

 

[Petrushka]

Dr David Griesinger did a study on the audibility of intermodulation distortions. His study was for loudspeakers, but since he had amplifier in the chain, he also looked at its effects. Here are his conclusion: (http://www.davidgriesinger.com/intermod.ppt)
View attachment 395431

And what he meant by "high quality amplifier".
View attachment 395432

Claims of audibility are rubbish without blind tests. You are asserting people can hear extremely low levels of high frequency noise when covered by much higher per levels of signal.

At worst, distortion products at those frequencies would slightly alter the timber of instruments. The instruments themselves would have to be producing something close to sine waves.

I say snake oil.

 

[Petrushka]

I was writing about the Ultra (a little bit off topic of this review, I admit).
We could hope for a more powerful and complete version, a sort of Ultra+Pro amp on steroid.
As it is, this Wiim pro is almost perfect, what ask for more in terms of esthetics, power, performance, ease of use, for this price ?

The Ultra is better than the Pro for what it can do, but IMO, at the same price level, it's not as "perfect" as the Pro integrated is : lack of Airplay is a serious miss, and the improvements over the other models in preamp, streaming and connectivity have not been pushed as far as it could have been done for a reasonable increase in price.

My guess is many costumers would prefer to pay 600 for an even more complete and better unit, a "real Ultra", than 400 for the released model.

If you were writing about the ultra, I was not responding to you. I was responding to people who think a bit of distortion above 15kHz makes the Pro a toy amp.

 

[kyuu]

I do not like this amplifier and would not give it my recommendation.

4) power cube behaviour is poor

Where are you seeing power cube results?

 

[Ajax]

Of course, this amp isn't purposed for listening in a large room with super expensive large low sensitivity speakers.
But, with a more sensitive and easy to drive speaker in a small or middle size room, 60 w/8 ohms could be fine, except for headbangers (it's not purposed for them too).

This is an excellent, nice looking, well engineered and built streamer/amp for people on a budget and more.

I hope Wiim will release in a next future a sort of Super Ultra Integrated with one balanced input, about 150W-200W on 8ohms power amp, a better and more versatile headphone output than the Ultra's one and, this time, Airplay connectivity too. The whole for, let's say, around 1500 €.

This would be "the perfect device on the market" without bleeding dry the wallet of sound enthusiasts, including people owning Revel Salons, Kef Reference or Magicos. Why spend more on electronics ? Most of the time, it's useless.

But, for the moment, i'm only dreaming. At present, a combination of Wiim Ultra plus a good powerful and cheap class D amp (3E Audio, Icepower, Hypex) could be just fine, just a little less complete and elegant solution.

I have always liked the front end, Streamer / DAC / DSP and the Power amp to be kept seperate. For me it just makes sense as you get better overall performance for the money and an easier upgrade path. Just add speakers.

SOTA performance for US$1,500 = minidsp Flex + Apollon / Hypex NCx500ST amp

Great performance for US$1,000 = Wiim Pro Plus + Buckeye / Hypex 502 amp

Even my old home theatre gear, Rotel AV 1066 pre + 1075 power, has been this way.

This is really inexpensive gear relative to the stuff you see in magazines such as Stereophile and Absolute Sound. I don't get the need for integrated amps and why they are sort after, two boxes instead of one is just not a big deal to me.

 

[Tell]

With 11-bits of IMD that's debatable.
IMD is the the most audible kind of distortion,that needs searching.

And how audible is that with real world music? I don't think it'll be very audible at all.

 

[NTK]

Claims of audibility are rubbish without blind tests. You are asserting people can hear extremely low levels of high frequency noise when covered by much higher per levels of signal.

At worst, distortion products at those frequencies would slightly alter the timber of instruments. The instruments themselves would have to be producing something close to sine waves.

I say snake oil.

Know that David Griesinger is the recipient of the Wallace Clement Sabine Medal from the Acoustical Society of America, for his "contributions to the understanding of human perception of sound".

 

[Petrushka]

Know that David Griesinger is the recipient of the Wallace Clement Sabine Medal from the Acoustical Society of America, for his "contributions to the understanding of human perception of sound".
View attachment 395497

Credentials are not audible, and just noticeable difference is not the same as important.

 

[NTK]

Credentials are not audible, and just noticeable difference is not the same as important.

I see that you have determined not to learn from the best in the business.

 

[Petrushka]

I see that you have determined not to learn from the best in the business.

• Amplifier distortion can produce distortion products below 20kHz that are audible (with difficulty) in the absence of other signals below
20kHz. - But with a high quality amplifier these distortion products are not audible in the presence of even
extraordinary ultrasonic sources such as rattling keys. Unless the amplifier is driven into clipping.

My takeaway is that such products are unimportant.

Besides, the link to his site is dead.

 

[ryanmh1]

I do not like this amplifier and would not give it my recommendation.

1) load dependency is unknown as it was measured only with resistive load
2) distortion rises with frequency too much
3) IMD 19+20kHz distortion is very poor, the odd order skirts are most likely a case of output inductors non-linearity
4) power cube behaviour is poor

Another toy instead of a serious amplifier.

Agreed, and I don't know that it's a known fact yet that taking the feedback after the output inductor necessarily solves everything that was seen in your excellent set of posts and measurements on the Aiyima A07. Obviously, it helps, in that the feedback can deal with some of the nonlinearity of the output filter, and measured distortion is indeed reduced. This paper https://nottingham-repository.worktribe.com/file/1009463/1/yutancoxgoh.pdf gives a fairly good overview of the IMD problems, and describes how the output filter potentially does make a PWM Class D worse than the "theory" would show. Unfortunately, it does not show that taking feedback after the filter necessarily solves it (it is silent on the issue).

There are other good papers too, but this one specifically finds (as you did also) that it does not necessarily take a 19+20kHz to excite the IMD issues in a typical closed loop Class D PWM amplifier, which commonly criticized as being "unrealistic". IMD is excited in Class D PWM amps by very, very benign signals. Not just torture tests. You found the same thing on a modified SMPTE test with 2kHz and a tiny 17kHz signal added--a shocking .26%! How much of it does PFFB solve? We don't know. If you have time hopefully you'll be able to do that on one of these PFFB variants, too! Hopefully solved, but who knows, since the Wiim Amp Pro with PFFB measures more or less the same on most potentially relevant measurements as the original.

Oh well, I think we've done enough clogging up a review thread with technical issues! I think as ASR's audience has broadened, this sort of discussion is less welcome and engaged in than it used to be. I've had enough of the "Prove you can hear it, bub" stuff. I can't, without assembling a panel of 30 people. IMD is possibly too hard to characterize and has too many causes and attendant "distributions" of the IMD products to do that, and make a blanket statement that "X" level is audible across the board. I think we just don't like it, that even cheapo Denon receivers have less of it, and will leave it at that.

 

[JeremyFife]

Agreed, and I don't know that it's a known fact yet that taking the feedback after the output inductor necessarily solves everything that was seen in your excellent set of posts and measurements on the Aiyima A07. Obviously, it helps, in that the feedback can deal with some of the nonlinearity of the output filter, and measured distortion is indeed reduced. This paper https://nottingham-repository.worktribe.com/file/1009463/1/yutancoxgoh.pdf gives a fairly good overview of the IMD problems, and describes how the output filter potentially does make a PWM Class D worse than the "theory" would show. Unfortunately, it does not show that taking feedback after the filter necessarily solves it (it is silent on the issue).

There are other good papers too, but this one specifically finds (as you did also) that it does not necessarily take a 19+20kHz to excite the IMD issues in a typical closed loop Class D PWM amplifier, which commonly criticized as being "unrealistic". IMD is excited in Class D PWM amps by very, very benign signals. Not just torture tests. You found the same thing on a modified SMPTE test with 2kHz and a tiny 17kHz signal added--a shocking .26%! How much of it does PFFB solve? We don't know. If you have time hopefully you'll be able to do that on one of these PFFB variants, too! Hopefully solved, but who knows, since the Wiim Amp Pro with PFFB measures more or less the same on most potentially relevant measurements as the original.

Oh well, I think we've done enough clogging up a review thread with technical issues! I think as ASR's audience has broadened, this sort of discussion is less welcome and engaged in than it used to be. I've had enough of the "Prove you can hear it, bub" stuff. I can't, without assembling a panel of 30 people. IMD is possibly too hard to characterize and has too many causes and attendant "distributions" of the IMD products to do that, and make a blanket statement that "X" level is audible across the board. I think we just don't like it, that even cheapo Denon receivers have less of it, and will leave it at that.

Darn, now I'm going to have to read the paper and actually think. Thanks for the link

 

[amirm]

Where are you seeing power cube results?

 

[quattro98]

I don't have a technical background in this field, but does this apply to self-oscillating designs like those from Hypex and Purifi?

Agreed, and I don't know that it's a known fact yet that taking the feedback after the output inductor necessarily solves everything that was seen in your excellent set of posts and measurements on the Aiyima A07. Obviously, it helps, in that the feedback can deal with some of the nonlinearity of the output filter, and measured distortion is indeed reduced. This paper https://nottingham-repository.worktribe.com/file/1009463/1/yutancoxgoh.pdf gives a fairly good overview of the IMD problems, and describes how the output filter potentially does make a PWM Class D worse than the "theory" would show. Unfortunately, it does not show that taking feedback after the filter necessarily solves it (it is silent on the issue).

There are other good papers too, but this one specifically finds (as you did also) that it does not necessarily take a 19+20kHz to excite the IMD issues in a typical closed loop Class D PWM amplifier, which commonly criticized as being "unrealistic". IMD is excited in Class D PWM amps by very, very benign signals. Not just torture tests. You found the same thing on a modified SMPTE test with 2kHz and a tiny 17kHz signal added--a shocking .26%! How much of it does PFFB solve? We don't know. If you have time hopefully you'll be able to do that on one of these PFFB variants, too! Hopefully solved, but who knows, since the Wiim Amp Pro with PFFB measures more or less the same on most potentially relevant measurements as the original.

Oh well, I think we've done enough clogging up a review thread with technical issues! I think as ASR's audience has broadened, this sort of discussion is less welcome and engaged in than it used to be. I've had enough of the "Prove you can hear it, bub" stuff. I can't, without assembling a panel of 30 people. IMD is possibly too hard to characterize and has too many causes and attendant "distributions" of the IMD products to do that, and make a blanket statement that "X" level is audible across the board. I think we just don't like it, that even cheapo Denon receivers have less of it, and will leave it at that.

 

[kyuu]

Ah ok, I was thinking they had seen the actual cube somewhere. Somehow it didn't strike me that this is equivalent information just presented differently.

But now I'm wondering what he means by the power cube behavior being poor. As long as you're not running speakers that get down to 2 ohms, seems ok?