• Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

DJM ACTIV Audio Ethernet EMI Filter Review (Bonus!)

Rate this Ethernet EMI Filter:

  • 1. Poor (headless panther)

    Votes: 106 67.9%
  • 2. Not terrible (postman panther)

    Votes: 33 21.2%
  • 3. Fine (happy panther)

    Votes: 16 10.3%
  • 4. Great (golfing panther)

    Votes: 1 0.6%

  • Total voters
    156
Ignorance is bliss . . in your case.
Ethernet is an error-corrected digital signal. Cables are twisted pairs (already very resistant against EMI) and you can get single or double shielded ones.

Apart from ground loops or packet losses (-> clear drop-outs) in your setup, there can be no audible difference when using Ethernet. The packets arrive bit-perfect, or they don't/error out. If you happen to have packet losses, get a better shielded cable for 20 $ and check your router settings.

Supposed "EMI filters" for a technology spefically designed to be resistant to EMI are 100% snake oil and any network tech or admin will be rolling on the floor laughing if you show them these devices. Feel free to ask one.
 
Steve McNally, an engineer behind DJM Electronics (and the makers of this Activ Audio filter) wrote a thought provoking article on his company's blog: Missing numbers: why smart people buy questionable products.

He makes no claims for audible improvement. He outlined where the filter might be useful. He explained how the filter works and what it is for. And he takes a nice swipe at his snake oil competitors and laughs at the subjective reviews of his product. Amir's review is mentioned but not explicitly named.

My take-away from his article: if I ever decide to set up an audio system next to an Arc welder and use both simultaneously, I might benefit from purchasing a filter.
 
My take-away from his article: if I ever decide to set up an audio system next to an Arc welder and use both simultaneously, I might benefit from purchasing a filter.
Lol, maybe if you are a heavy metal fan?
 
I re-read that article again (it is doing its rounds in subjectivist discussion, e.g. Michael Lavorgna's article here). Anyway it is clear from reading that article that they are a manufacturer of industrial EMI filters for legit reasons, e.g. military, MRI, and heavy industry where EMI interference of ethernet signals may be a legitimate concern. They entered the audiophile market by accident:

1791423431734.png


So this product that Amir reviewed is essentially one of their industrial filters but repackaged in a pretty box. He admits as much here:

1791423565696.png


The rest of the article discusses the engineering challenges of EMI filtering, makes claims about the effectiveness of their technology, and how to evaluate the performance of a product. NOT ONCE does he claim that it would improve the sound of your hifi system, and in fact he argues against it:

1791423803879.png


Early in the article, he does speculate (but not prove or test) a possible mechanism through which EMI filtering might improve the sound:

1791424700821.png


I have no idea whether he is right or not. He does complain about the methodology used by @amirm, noting that Amir failed to test the EMI noise rejection capability of the filter. I am not sure how Amir could create EMI noise, I doubt if he has an MRI machine at home. Perhaps they could have suggested a better way to conduct the test. It would certainly be interesting to inject EMI noise into Ethernet and see if it makes it to the DAC output (and we are NOT talking about packet loss which would surely manifest itself in a way nobody would argue with).
 
I think this isn’t about noise making it to the DAC, it’s about Ethernet no longer working, and the filter fixing it. And yes, this would be audible ;)
 
I think this isn’t about noise making it to the DAC, it’s about Ethernet no longer working, and the filter fixing it. And yes, this would be audible ;)

That's what the device is for, yes. But the audiophile claim is that ethernet noise manifests itself in DAC output. This is what many of Amir's tests are about - debunking audiophile tomfoolery. As McNally says, plausibility is not proof.
 
Never have I ever heard those audiophiles complain about audible noise, and then one of these devices fixes that. It's always about darker backgrounds, a wider soundstage, etc... Claiming that audible noise is gone is one of the easiest things to prove; you don't even need to play any audio for that.
 
Last edited:
He does complain about the methodology used by @amirm, noting that Amir failed to test the EMI noise rejection capability of the filter
But I didn't. It is in the review where I induced a magnetic field:

index.php


Result was measurable but at an incredibly low -140 dB. Reason is that there is so much isolation between Ethernet input to a streamer and its output, as it should be.
 
But I didn't. It is in the review where I induced a magnetic field:

index.php


Result was measurable but at an incredibly low -140 dB. Reason is that there is so much isolation between Ethernet input to a streamer and its output, as it should be.

Actually you tested it outside of it's published spec which they state is 10KHz to 40GHz
 
Ethernet is an error-corrected digital signal. Cables are twisted pairs (already very resistant against EMI) and you can get single or double shielded ones.

Curious, what do you think of what Gemini said below:

Over 80% of generic patch cords fail basic TIA/ISO category specifications because the twisted pairs are untwisted too far back before being crimped into the plug. Untwisting pairs right at the connector destroys the balanced geometry, producing massive near-end crosstalk (NEXT).

  • Blue Jeans terminates their cables by hand in Seattle using dimensional crimp dies, trimming the untwisted length to the absolute physical minimum.
  • Crucially, every single Blue Jeans patch cable is individually tested and certified on a Fluke DTX/DSX CableAnalyzer before it leaves the bench, shipping with an individual printed certification report detailing exact return loss and crosstalk margins.

The Audible Result​

Generic cables inject high-frequency parasitic ringing and ground-plane noise that bleeds past transformer interwinding capacitance, producing the classic "digital" glare, thin tonality, and flat soundstage.

Blue Jeans cables behave as a textbook-ideal, 100-ohm balanced transmission line: zero ground-coupling paths, virtually no signal reflections, and an unperturbed PHY layer. That is why they sound unforced, dynamically open, and completely free of artificial grain.
 
Curious, what do you think of what Gemini said below:

Over 80% of generic patch cords fail basic TIA/ISO category specifications because the twisted pairs are untwisted too far back before being crimped into the plug. Untwisting pairs right at the connector destroys the balanced geometry, producing massive near-end crosstalk (NEXT).

  • Blue Jeans terminates their cables by hand in Seattle using dimensional crimp dies, trimming the untwisted length to the absolute physical minimum.
  • Crucially, every single Blue Jeans patch cable is individually tested and certified on a Fluke DTX/DSX CableAnalyzer before it leaves the bench, shipping with an individual printed certification report detailing exact return loss and crosstalk margins.

The Audible Result​

Generic cables inject high-frequency parasitic ringing and ground-plane noise that bleeds past transformer interwinding capacitance, producing the classic "digital" glare, thin tonality, and flat soundstage.

Blue Jeans cables behave as a textbook-ideal, 100-ohm balanced transmission line: zero ground-coupling paths, virtually no signal reflections, and an unperturbed PHY layer. That is why they sound unforced, dynamically open, and completely free of artificial grain.
I asked Gemini ...
1791936799679.png
 
Gemini should run for politics. Says stupid things, and then denies saying said stupid thing.
 
Gemini should run for politics. Says stupid things, and then denies saying said stupid thing.
Like politicians, you can have LLMs say anything you like if you feed them enough weasel words ;) A simple bit of prompt engineering will give you exactly the output you like, facts be damned!
 
Back
Top Bottom