After last FW upgrade, Wiim seems to be no more bit perfect.

merifon

Senior Member
Joined
Oct 25, 2022
Messages
446
Location
Bulåggna
I would like to point out that some users refer that bit perfect reproduction via toslink verification tests on RME ADI-2 and other gears, is lost. Deviations are especially for anything over 44.1/16.
Can someone verify and provide?
Bit perfectness was the main reason to use wiim...

thanks in advance
 
Look for "60hz" in the Amazon music. When playing back 60Hz tone, distortion can be observed at a fixed volume.
 
I have the same issues with my device so I hope a fix will come very soon.
We will use the older firmware version and new firmware version to test the sound quality with AP. In the meantime, can you help to send us the MAC address of your device? We will downgrade your device to the 20220922 firmware version and please help to check if the distortion still persists.
 
We will use the older firmware version and new firmware version to test the sound quality with AP. In the meantime, can you help to send us the MAC address of your device? We will downgrade your device to the 20220922 firmware version and please help to check if the distortion still persists.
Of course
 
Last edited:
With the current firmware 4.6.434158, I performed some tests:

Test in Adobe Audition entirely in digital domain: original generated tone of 1000 Hz at 0 dB (left) then increased the volume with only 0,3 dB, saved the file and opened again (right). See the distortions (harmonics) appeared because of clipping over 0 dB.

1667459826508.png
1667459915892.png



This test shows exactly the behavior of Wiim Mini - source was the clean 0 db file and I captured similar distortion when casting to Wiim (see below capture).
1667460567063.png


I interpret this as there is some type of processing in Wiim which exceeds 0 dB, therefore all the distortion and harmonics.
 
I got the 22/09 firmware 4.6.430138. I repeated the tests and:

- The 0 dB sample still produces distortions
- Anything below does not produce distortions
- No more 'limiting' e.g. you see differences in output level now (versus no difference between 0 to -1,1 dB)

So, I think only partly solved... because still there is something at 0 dB.

Entire test pattern analysis:
1667468707775.png



At 0 dB:
1667468853213.png



At -0,3 dB:

1667469097877.png



At -0,7 dB:

1667469045620.png
 
Great to know also your feedback.

I still don't understand the harmonics at 0 dB, with the 1000Hz tone test. Maybe that is a different issue in the DAC of the WiiM.

Otherwise the results we have are consistent below 0 dBFS.
 
Great to know also your feedback.

I still don't understand the harmonics at 0 dB, with the 1000Hz tone test. Maybe that is a different issue in the DAC of the WiiM.

Otherwise the results we have are consistent below 0 dBFS.
in analog domain, levels are relative. Vcc, OP amp gain and other things can easily lead to not perfectly alignement to level that is produced by a digital 0dB signal. In such domain, make sense to lower the overall volume to get some headroom. Generally speaking a single 5Vcc power, doesn't help in such way.
 
Back
Top