Difference between WiFi and Ethernet

[Musician]

When on, the wifi inside the WiiM should contaminate all the analog parts inside the WiiM with RFI.
If you dont believe what RFI can be responsible for when using analog audio , try making a 1 meter RCA cable with only two wires and no shielding. It gonna sound different, brighter, and slightly worse than a shielded cable .

This is a very interesting seminar about RFI contamination with Rob Watts , Chord electronics.

For serious listening and not using a separat dac with a shielded metal cabinet, I would not recommend using wifi . There are easy solutions with ethernetcables, such as using powerline adapters.

Because of this risk of RFI issue, I have high hopes for the new WiiM ultra that has a shielded metal case, that might bring even better sound when using the inbuilt DA converter.

 

[Jls]

CDs also have only one bit error detection.

I’ve spent a lot of time ripping CDs with Exact Audio Copy. Most of my disks are purchased used and are not perfect. Since CDs don’t have error correction, you have to make choices about how to handle errors. In its paranoid mode, EAC does a CRC for each track and compares it to what other users got for the same track.

In other modes it simply retries a sector until it gets an error free read.

What real time playing does is either silence the bad sector, or fill in from previous and subsequent sectors. We’re are typically talking about a millisecond.

 

[Jls]

Ethernet cables are unshielded, which is another source of fun.

 

[Musician]

Ethernet cables are unshielded, which is another source of fun.

But its carrying digital signals ( but your right its in analog shape in its traveling ) . The ethernet might as you say be contaminated with a lot of RFI , but the impact after the DA conversion will be less in most cases, compared to using wifi.

However, there are some debate by audiophiles about digital signals traveling in an analog way ( like with ethernet cables ) might be to noisy if the DA cant recover it in a good way - thus impact the analog sound in a bad way.

We must also remember there are an industry that will tell us there is expensive solutions on problems that might not exist.

 

[FreakyKiwi]

Not to say that none of this is fun and interesting, my opinion here is that whatever improvement any of this may make, given the inherent reliability of the network stack, it is surely incredibly small compared to other factors more easily within our control.

I would need to have exhausted all other avenues for improvement unless my network connection was so bad that is was causing noticible issues and even in that case would only look to replace a switch, cable or access point.

This is just my opinion based on my own experience and no testing to back it up, so I mean no offence to anyone with a different experience.

 

[simbun]

In its paranoid mode, EAC does a CRC for each track and compares it to what other users got for the same track.

That technology is called AccurateRip and is available in all modes. CUETools has a similar technology but is also able to fix ripping errors if enough matching discs have been submitted.
I run in Burst mode ~95% of the time knowing that a match to AccurateRip is proof of a bit perfect rip.

 

[Wiimer]

Because of this risk of RFI issue, I have high hopes for the new WiiM ultra that has a shielded metal case, that might bring even better sound when using the inbuilt DA converter.

A simple layman's question: how can a metal case be expected to have a shielding effect even though the Wi-Fi antenna resides inside the device?

 

[Musician]

Lmao, I love how @FreakyKiwi can read and summarise months of testing, discussions, formulation of falsifiable hypotheses etc... all in the space of 13 minutes! Mighty impressive who needs chatGPT?? We should get the guy to do the same with the cancer literature. He'll have it all cured and in our rear-view mirrors by next weekend?!

And thanks to @Jls too, I appreciate that you are trying to share your opinion in a cordial manner.
Unfortunately some of us think differently about audio - in short - bits are not bits the same way it is for an app download. If you go into details about how DACs function etc then you might uncover something that will open your line of thinking to other possibilities. Things like common mode interference for example. But I respect your choice if you feel have already got enough information to form a fixed opinion...

Just to elaborate on what you said too - yes error correction and buffering exist in the digital domain (mainly in asynchronous data transmission like Ethernet and USB). But there is no error correction in S/PDIF (only single bit error detection) and no buffering both over coax or optical...


Yes a few people have gone that route. It tends to work out well for them. However I went the air-gap route with a WLAN repeater in bridge mode that the lads on Ethernet sound proposed and tested... overall the cheapest/easiest to implement

There is always buffering at the receiving end of the dac nowadays ( at the input receiver of the dac ) when using spdif, both with toslink and electrical spdif. This is a problem from the past, 20 years ago.

 

[Jls]

But its carrying digital signals ( but your right its in analog shape in its traveling ) . The ethernet might as you say be contaminated with a lot of RFI , but the impact after the DA conversion will be less in most cases than using wifi.

However, there are some debate by audiophiles about digital signals traveling in an analog way ( like with ethernet cables ) might be to noisy if the DA cant recover it in a good way - thus impact the sound in a bad way.

The digital signal in wifi can have dropouts if the router is flaky, the receiver is inadequate, or the distance too great. Dropouts are entirely a separate issue from fidelity. The character of sound cannot be altered by the mode of transmission with an error correcting protocol.

Now, once the signal becomes analog, within a device, it is susceptible to interference, and that is why we have engineers. I would hope that expensive devices have better isolation within the box than cheap devices, but I would not take this for granted. If the marketing of a device includes snake oil, I’m not buying.

Anyway, there’s an arcane technical term for interference. It’s called noise. It’s measurable. Send a file like Phillip Glass’s “4:33”and turn the volume up. If you hear anything, it’s noise created by the DAC or by the analog components in the amplification chain.

 

[Musician]

A simple layman's question: how can a metal case be expected to have a shielding effect even though the Wi-Fi antenna resides inside the device?

Theres still problems with wifi if there is RFI bouncing around inside a shielded metal cabinet containing analog parts after the dac.

My high hopes for the WiiM ultra was purely when using ethernet and the inbuildt DAC . Because of the shielded cabinet with the ultra, the ultra might sound slightly better than the pro plus when using the internal dacs.
Maybe Im wrong and they will sound the same, but time will tell.

 

[harkpabst]

I changed to cable to stop drop outs. I was not hoping for improvements.

If you had drop outs before using Wi-Fi then that connection was definitely not stable. Most probably there have been interruptions you didn't even recognise as drop outs, but just inferior sound quality. Mystery solved.

There are easy solutions with ethernetcables, such as using powerline adapters.

Powerline adapters do contaminate the electrical supply for all devices by their working principle.

 

[Jls]

I’m going to have to correct myself. Glass’s 4’33” is not silence. Any performance includes the ambient sounds in the room or auditorium, including the audience. Sill a good test of equipment and recording technique.

Actually, probably a better than average test.

 

[MKUltra]

Ethernet cables are unshielded, which is another source of fun.

Who told you that?

Here, go have a quick browse, no need to read, just observe the photos: https://www.universalnetworks.co.uk/faq/what-does-utp-s-utp-ftp-stp-and-sftp-mean/

 

[Jls]

Who told you that?

Here, go have a quick browse, no need to read, just observe the photos: https://www.universalnetworks.co.uk/faq/what-does-utp-s-utp-ftp-stp-and-sftp-mean/

 

[Jls]

I overgeneralized. But shielded Ethernet is effective if the couplers and sockets are shielded and grounded. Using shielded cable into plastic sockets is not particularly effective.

 

[willbewill]

I won't use Power line adapters to 'wire' remote ethernet - they can (but not always it seems) be a great way of introducing 'noise' to your power lines especially if on the same ring main as your hifi. I use actual CAT wired connections.

 

[BowsAndArrows]

There is always buffering at the receiving end of the dac nowadays ( at the input receiver of the dac ) when using spdif, both with toslink and electrical spdif. This is a problem from the past, 20 years ago.

well, if you're referring to FIFO buffering + ASRC and PLL/PPLL masterclock recovery. yes these measures can reduce S/PDIF jitter. however none can eliminate it entirely, and i believe the results are heavily dependent on the specific implementation. i could be wrong though...

 

[Jls]

There has to be something wrong here, possibly with my understanding. With a digital source, such as streaming, the incoming data is asynchronous, it just doesn’t matter what’s happening with the original clock. The incoming data will be error corrected, buffered, and fed to the DAC. There is only the local clock.

Update to add:

Due to buffering, there is no informational difference between streaming and downloading, except that an interruption longer than the buffer will cause dropouts.

If the device is susceptible to RF interference internally, that cannot be corrected by cabling. But it will be measurable.

 

[VintageFlanker]

Blind tested my partner

How did you do that ?

 

[Smartplug]

I had the Wimm for over a week now
Yesterday I finally connected the Wiim via an Ethernet cable.
More clarity
Blind tested my partner as she also agrees more clarity. To be honest was not expecting any difference, but glad I did.

Don’t forget to buy audio quest cinnamon to complete your golden listening. It just hilarious when someone claim such thing without proof of audio precision result.