WiiM Mini Streamer

[onlyoneme]

That's somewhat weird considering @morillon still observes distortions, I can still see bad waveforms and @JimmyMac had MQA pass-thru issues which has been fixed by FW rollback. My FW didn't change.

 

[onlyoneme]

I've asked WiiM to rollback my FW to 430138 version.

 

[onlyoneme]

me
I'm just waiting for someone to prove me wrong... no worries...
or that my wiim has broken down...?
Honestly, I don't take pride in it.
do it..
;-)

firm 432755
home 221018

but I think the case since previous firmwares..

I'm really sorry but I don't have an equipment nor necessary knowledge to be able to repeat your tests...

 

[onlyoneme]

I've repeated my test with analog output to be sure it's not toslink out-in or optical cable related. Unfortunately still the same issue.

 

[Merifon]

I've repeated my test with analog output to be sure it's not toslink out-in or optical cable related. Unfortunately still the same issue.

How do you capture the output of Flex? Is another PC/sound interface or is the same running the test signal?
It seems that in the USB case, some sync between sample points time is kept, whilst in toslink case, samples are not time synchronized and sample point can or cannot take the peak of impulse.

 

[onlyoneme]

How do you capture the output of Flex? Is another PC/sound interface or is the same running the test signal?
It seems that in the USB case, some sync between sample points time is kept, whilst in toslink case, samples are not time synchronized and sample point can or cannot take the peak of impulse.

Flex is able to send its output to the PC over USB interface for any input used as it is both AD and DA converter. I capture it with ocenaudio using ASIO drivers.

Waveforms look very similar in case of WiiM for both toslink and analog outputs.

 

[Turambar]

Qobuz, which is otherwise reduced to a 1990s throwback experience through bubble upnp or mconnect.

You can run Qobuz from the WiiM Mini streaming application, with a look and feel pretty similar to the Qobuz app itself.

 

[Phorize]

You can run Qobuz from the WiiM Mini streaming application, with a look and feel pretty similar to the Qobuz app itself.

Does it give access to the wider features of the app? My experience of other software using their api is that half of the playlists views so are missing, as are all of ghe articles etc.

 

[Turambar]

Does it give access to the wider features of the app? My experience of other software using their api is that half of the playlists views so are missing, as are all of ghe articles etc.

I think so, though I'm not using 100% of the features of the app.

It works fine for me for months now.

Take a look for yourself.

 

[Merifon]

Flex is able to send its output to the PC over USB interface for any input used as it is both AD and DA converter. I capture it with ocenaudio using ASIO drivers.

Waveforms look very similar in case of WiiM for both toslink and analog outputs.

So, the same usb and therefore the same PC, are involved in the in and out passes of that test. This could be explain the synchronized sample points (just a little process delay).
Maybe I'm wrong but a different device should be used to capture the output, possibly at higher sample rate than signal subject of test. The imbebbed wiim toslink clock cannot be in sync with the USB and is possible that sometimes it is sampled the very peak of signal an sometimes not.
Is just my opinion, since I don't know what flex really does...

 

[Music1969]

for you at 44k16bit ;-)

the actual distortion is in mode no eq at 44k to...
as soon as we approach 0dbfs


fast measurement
to this date
same protocol at -0.1dbfs 998hz 44k 16bits


recently (for example mid-September), the differences were minimal on such a basic measurement in toslink, with chromecast wiim, normal.... (just the wiim which has significantly more jitter than the chromecast)
but now

You are measuring analogue output distortions correct ?

 

[onlyoneme]

So, the same usb and therefore the same PC, are involved in the in and out passes of that test. This could be explain the synchronized sample points (just a little process delay).
Maybe I'm wrong but a different device should be used to capture the output, possibly at higher sample rate than signal subject of test. The imbebbed wiim toslink clock cannot be in sync with the USB and is possible that sometimes it is sampled the very peak of signal an sometimes not.
Is just my opinion...

There is the ASRC always involved on Flex side, so I guess it doesn't really matter that I send and receive signals on the same interface. Moreover WiiM's analog output is affected the same way so it cannot be something related to the spdif clock.

 

[Merifon]

it cannot be something related to the spdif clock.

Sorry, I was meaning wiim internal clock, regardless the kind of output.
I don't want to defend wiim but it's unknown how flex manages that return bus,under different cases.
Despite everything, it would be interesting to see what would happen to the signal produced by the PC and any other sources, when captured on flex analogue output by an external device.

 

[onlyoneme]

Sorry, I was meaning wiim internal clock, regardless the kind of output.
I don't want to defend wiim but it's unknown how flex manages that return bus,under different cases.
Despite everything, it would be interesting to see what would happen to the signal produced by the PC and any other sources, when captured on flex analogue output by an external device.

Maybe you're right and I'm completely wrong as I do not have any expert skills in the measurements area. Differences I can see still intrigues me so do those observations related to possible distortions. I'm going to repeat my tests after FW downgrade if possible.

 

[Jpdelaney.tml]

This is the lastest entry in Wiim Forum re SPDIF output distortion....."Regardless of RME test results I can still observe the same issues as before even if EQ is off."
Comments?

 

[Merifon]

Maybe you're right and I'm completely wrong as I do not have any expert skills in the measurements area. Differences I can see still intrigues me so do those observations related to possible distortions. I'm going to repeat my tests after FW downgrade if possible.

I'm not absolutely meaning that, it was generally speaking, any experimental results have to be confirmed by a different kind of test.
For exemple, if missing a separate capture system, trying a different toslink source, to be compared...

 

[onlyoneme]

Some THD measurements made in REW for 96 kHz sweep signals, vertical scale in dBFS:

Does it look good? I don't think so.

 

[Music1969]

Sorry, I was meaning wiim internal clock, regardless the kind of output.
I don't want to defend wiim but it's unknown how flex manages that return bus,under different cases.
Despite everything, it would be interesting to see what would happen to the signal produced by the PC and any other sources, when captured on flex analogue output by an external device.

Agree with this

 

[Merifon]

Some THD measurements made in REW for 96 kHz sweep signals, vertical scale in dBFS:


Does it look good? I don't think so.

No, they don't look good, even for usb should be better.
If the test paths and methods are the same as in the previous images, it is confirmation that further investigations should be carried out.

 

[Music1969]

even for usb should be better.

Agree - this is proof your measurement method is sub-optimal @onlyoneme

So far, RME engineers are still winning ;-)

I will do some analysis in a few weeks, as promised previously