WiiM Pro - Review & Measurements (Streamer)

[magreen]

cheapaudioman has a new video in which he vaguely alleges without evidence that the Wiim Pro and Ultra digital outputs suffer from digital jitter to a degree that is audibly problematic. He says you can clean it up only on the Wiim Ultra because it has USB output you can run through a DDC to reclock. The video is all about using an SMSL DDC for this. He admits he doesn't understand the jitter issue, and relies on another youtuber, OldGuyHifi, for the allegation that Wiim suffers from jitter problems. And he claims it's totally audible. I'm skeptical. The jitter allegations seem to 100% contradict the very low jitter measurements for digital outs in the OP of this very thread.

Thoughts from people much more knowledgeable about audio than me?

Here's the link to his new video:

 

[Jimbob54]

cheapaudioman has a new video in which he vaguely alleges without evidence that the Wiim Pro and Ultra digital outputs suffer from digital jitter. He says you can clean it up only on the Wiim Ultra because it has USB output you can run through a DDC to reclock. The video is all about using an SMSL DDC for this. He admits he doesn't understand the jitter issue, and relies on another youtuber, OldGuyHifi, for the allegation that Wiim suffers from jitter problems. And he claims it's totally audible. I'm skeptical. The jitter allegations seem to 100% contradict the very low jitter measurements for digital outs in the OP of this very thread.

Thoughts from people much more knowledgeable about audio than me?

Here's the link to his new video:

Seems to me like a ploy to a) get views and b) aid SMSL in selling "cheap" wonderfixes for non-existent issues.

I would ignore absent of any actual evidence either from measurements or properly conducted (blind) listening tests.

 

[Matias]

@magreen See here.

Jitter over digital outputs for different WiiMs

Let's talk about the jitter over digital outputs. I will use the Holo Audio Spring 3 DAC which is a true NOS DAC and has an unique ability to turn off its PLL. This way I can compare a raw performance of the outputs in j-test. Ultra, toslink: Ultra, coaxial: Pro, toslink: Pro...

forum.wiimhome.com

 

[Mat]

cheapaudioman has a new video in which he vaguely alleges without evidence that the Wiim Pro and Ultra digital outputs suffer from digital jitter.

Of course it has jitter. It aint magically immune.

 

[magreen]

Of course it has jitter. It aint magically immune.

Ok but the allegation is that the jitter is audibly problematic. I’ve edited my post accordingly.

 

[davidc]

Why would you buy the low cost WiiM stuff, and then spend another 20% on an add-on.

 

[magreen]

@magreen See here.

Jitter over digital outputs for different WiiMs

Let's talk about the jitter over digital outputs. I will use the Holo Audio Spring 3 DAC which is a true NOS DAC and has an unique ability to turn off its PLL. This way I can compare a raw performance of the outputs in j-test. Ultra, toslink: Ultra, coaxial: Pro, toslink: Pro...

forum.wiimhome.com

Interesting thread. Those measurements also seem to contradict the digital jitter levels measured in the OP in this thread.

 

[kyuu]

Not only in this thread's OP, but all of Amirm's measurements show no jitter issues with any of the Wiim devices on their digital outputs. This is a non issue.

 

[Limopard]

Not only in this thread's OP, but all of Amirm's measurements show no jitter issues with any of the Wiim devices on their digital outputs. This is a non issue.

There is always someone who has to kick a can down the road. (or as we Germans say: "eine neue Sau durchs Dorf treiben")

 

[VintageFlanker]

Ok but the allegation is that the jitter is audibly problematic

It's not.

 

[nanook]

Not only in this thread's OP, but all of Amirm's measurements show no jitter issues with any of the Wiim devices on their digital outputs.

@kyuu Maybe I missed something - correct me if I'm wrong.
Did Amir indeed measure jitter on the digital outputs? There's the usual jitter spectrum measured in the analog domain in Amir's review, but I cannot recall measurements aside the ones presented (and linked in this thread) in the WiiM forum.
Very small jitter in the analog domain does not guarantee this holds true for the digital outputs. Limited bandwidth (e.g. cheap LED in the Toslink transmitter) can lead to heavy ISI (Inter-Symbol-Interference).
And e.g. this one (WiiM Mini, Toslink , 48kS/s according to the WiiM forum thread linked in post #867 above) is far from looking nice: https://forum.wiimhome.com/attachments/1722433737842-png.9694/

If you are actually saying "... no jitter issues ..." in terms of audibility I do agree, but from a circuit development and measurement point of view -95dB definitely are not state-of-the-art. And we are discussing, comparing and criticizing a lot of shortcomings here in ASR that are very very far below getting audible.

 

[Pritaudio]

Would it work ( not necessarily better) with the usb port on the Wiim amp pro.
It could serve as having the features of the Wiim ultra but with an inbuilt amp ( if you decide you need the amp section later).
So, does the usb port on the Wiim amp pro support digital out for this to be suitable option.

 

[Pritaudio]

The guy hears a difference and fails to understand that it could be frequency response of partnering components.

 

[kyuu]

@kyuu Maybe I missed something - correct me if I'm wrong.
Did Amir indeed measure jitter on the digital outputs? There's the usual jitter spectrum measured in the analog domain in Amir's review, but I cannot recall measurements aside the ones presented (and linked in this thread) in the WiiM forum.

You're correct, Amir's are all for TOSLINK in and RCA out.

Very small jitter in the analog domain does not guarantee this holds true for the digital outputs. Limited bandwidth (e.g. cheap LED in the Toslink transmitter) can lead to heavy ISI (Inter-Symbol-Interference).
And e.g. this one (WiiM Mini, Toslink , 48kS/s according to the WiiM forum thread linked in post #867 above) is far from looking nice: https://forum.wiimhome.com/attachments/1722433737842-png.9694/

Pretty sure all TOSLINK outputs use cheap LEDs. That's kinda what it's designed for? It's not a super high bandwidth, long-distance interconnect. TOSLINK cables are plastic.

In any case, the OP shows that even on the Pro, if you're going digital-in to digital-out you're getting bit-perfect passthrough with no jitter issues even on TOSLINK:

With the Pro, the ADC leaves quite a bit to be desired but if you're going digital-in there's nothing to worry about.

If you are actually saying "... no jitter issues ..." in terms of audibility I do agree, but from a circuit development and measurement point of view -95dB definitely are not state-of-the-art. And we are discussing, comparing and criticizing a lot of shortcomings here in ASR that are very very far below getting audible.

OK, but we're not talking about achieving the state of the art in this context. This discussion started with the posting of a video that claims there is an audible issue with the digital-out that needs fixing by way of reclocking. If you want to argue that an audibly transparent $150 streamer (as long as you're not using a high-quality analog source) needs better jitter performance, then have at it I suppose.

 

[nanook]

Pretty sure all TOSLINK outputs use cheap LEDs. That's kinda what it's designed for? It's not a super high bandwidth, long-distance interconnect. TOSLINK cables are plastic.

Sure, its LED (not laser) and plastic multimode fibers with huge core diameters, but still there are significant differences in rise- and fall- times between different LEDs.

In any case, the OP shows that even on the Pro, if you're going digital-in to digital-out you're getting bit-perfect passthrough with no jitter issues even on TOSLINK:

Analyzing the sampled digital data of the digital output will not let jitter get visible - it's just digital data, the timing information is already removed by the sampling process.
EDIT: The spectrum is then calculated using the accurate time-points given by the calculation - not given by a real jittery clock.
And if you D/A- convert this data using a perfect clock, there's "no" jitter in the spectrum of the analog signal. This is what the PLL and re-clocking in the DACs aims at.
In a real system however this "perfect" clock has to adjust to the incoming bitstream, it needs a PLL. If the time constant of the loop filter is very low, the PLL may not be able to follow the varying incoming clock fast enough and loose its lock- state. So choosing the loop filter is a compromise and the PLL will thus suppress the incoming jitter only to a certain extend. Some DACs have a very good jitter suppression in their PLL even at low frequencies and others do not.

In short: To my understanding the above spectrum (digital out analyzed as "sampled digital data") does not tell anything about the jitter at the Toslink output.

OK, but we're not talking about achieving the state of the art in this context. This discussion started with the posting of a video that claims there is an audible issue with the digital-out that needs fixing by way of reclocking.

Absolutely agreed, that video is indeed vastly misleading also to my opinion. And last-not-least most modern DACs do have reasonable reclocking. I would actually love to see real jitter measurements (time domain) at the digital outputs (not feasible with the AP I'm afraid) and jitter suppression measurements for digital inputs of DACs.
From a circuit design point of view, achieving low jitter on digital outputs does not have to be expensive. I'm afraid the key to a good jitter performance is a good PCB artwork (apart from clean and stable supply voltage for the PLL and the whole clock path).
A nice example is the SMSL PO 100 PRO (measured in the WiiM forum for comparisson). It's USB bus powered and it manages to produce a very clean digital output.

 

[antcollinet]

Some DACs have a very good jitter suppression in their PLL even at low frequencies and others do not.

But (nearly) all of them can suppress jitter such that the resulting distortion is inaudible.

And I only say "nearly" because not all have been measured. I think all that have been measured can, and I fully expect there are no DACS currently on the market** which (assuming they can lock onto the signal at all) are not able to suppress the jitter sufficiently.

Jitter is basically a solved problem and has been for decades. The subject simply wastes pages and pages of discussion and time. It is the go to boogyman for audiophools who cannot accept that the veils they constantly hear being lifted are only being lifted by their perceptive biases, and not by any performance improvements in their latest and greatest upgraded gear.



**Except (perhaps) for some of the 10-a-penny junk items sold on ebay and amazon. Though even then I'm not aware of any that have been measured as having audible problems with jitter.

 

[nanook]

Jitter is basically a solved problem and has been for decades. The subject simply wastes pages and pages of discussion and time. It is the go to boogyman for audiophools who cannot accept that the veils they constantly hear being lifted are only being lifted by their perceptive biases, and not by any performance improvements in their latest and greatest upgraded gear.

No need to argue about this. I fully agree the amount of jitter shown for the WiiM Mini (linked in post #867) will not be audible, but certainly this is a spectrum we would rather see improved in the next redesign. From a circuit design point of view this amount of jitter is not necessary. The WiiM Mini has restricted space, so this migth be the reason. More on jitter later, although I agree "it's a solved problem".

Here at ASR we are discussing a lot things that are not audible; SINAD, THD, IMD (incl. the so called "ESS- hump"). All of this is very likely at inaudible levels for a decent 100€ DAC. If it's about audibility of measured parameters we could have shut down this forum before long. For me these are technical discussions for the sake of technical knowhow and product improvement - I'm an engineer .
Strictly I do not argue about audibilty.

But (nearly) all of them can suppress jitter such that the resulting distortion is inaudible.

I fully agree, but strictly speaking, we do know very little about the jitter suppression capabilities of DACs, because it's simply not measured in the regular reviews.
I assume the optical output signal of the Audio-Precision will be as good as enginieering allows with reasonable effort - it's an instrument. There will be lots of cost optimized optical (and even coaxial) outputs around that are a lot worse. Thus "all" we see in Amir's plots is the amount of jitter the DAC introduces itself when fed by a state-of-the-art optical signal.
Afaik the AP can introduce jitter onto the digital output signals and using an intentionally jittery signal with known frequency and TIE statistics it would be possible to assess the jitter suppression capability of a DAC as a function of frequency. But - correct me if I'm wrong - I have only seen very few measurements on this.

Another property not being measured is the jitter statistics of digital sources like streamers. This is what onlyoneme showed in the thread in the WiiM forum. He used a DAC where he apparently can more-or-less turn off the loop filter of the PLL and is thus capable to qualitatively assess the jitter at the digital input of the DAC by measuring the converted signal in the analog domain.
I do not expect many digital sources showing worrying levels of jitter - but we just don't know.

 

[JustJones]

Thus "all" we see in Amir's plots is the amount of jitter the DAC introduces itself when fed by a state-of-the-art optical signal.

Depends on the DAC and what inputs it uses, here's the SMSL D6s on all 3 inputs

 

[Ralph_Cramden]

cheapaudioman

 

[antcollinet]

but certainly this is a spectrum we would rather see improved in the next redesign

Personally I don't care. I'd rather manufactures spend their efforts on things that do impact the customer, rather than those that don't. Sure - if they can just fit a better LED/driver without spending weeks of R&D, or increasing the BOM cost - fine. But otherwise I'm not particularly worried.

I fully agree, but strictly speaking, we do know very little about the jitter suppression capabilities of DACs,

But what we do see in post #867 - is even with the PLL turned off, and jitter rejection not happening, the jitter is still not audible.

I'm removing myself from the discussion at this point to stop wasting yet more time, effort and pixels to discuss a non problem.