Regen Session: Part II
The evaluation session we held on Labor Day was of great interest according to the stats provided by the website. Clearly, there’s a lot of interest in the Regen device and its claim of fidelity enhancement. Some readers have commented and complained that we didn’t do a rigorous A|B|X test, but that wasn’t my intention. I wanted to do two things with Russ and the Regen unit. The first was to listen to the exact same system in the studio with and without the Regen in the signal path. Frankly, I was curious. I’ve been strongly criticized by Michael over at Audiostream about my willingness to proclaim audiophile accessories “snake oil” without actually experiencing them myself.
I countered by saying there are some audiophile tweaks that simply don’t pass the smell test…at least for me. The “Teleportation Tweak”…and frankly all of the other enhancements offered by Geoff Kait (who is a self described, unabashed “snake oil” salesman)…is a prime example of something that is so ridiculous that I don’t feel compelled to spend $60 on a phone call to test it. But OK, Michael made his point. He’s a believer and I had an opportunity to test the Regen without having to invest in the box, so I stepped up. Russ and his friend from the audiophile society were very generous to supply a unit and bring some other equipment along to do what I consider a very fair and balanced evaluation.
I’ve already written the results of that listening. It certainly won’t end the debate among believers and non-believers but I’m glad that we spent the time to do what we did. Chip and Russ believe that they heard a change in the sound coming from my monitors and I heard no change at all. I mean absolutely no change…nothing. They admitted that the change was very subtle and barely noticeable. I found the box didn’t enhance the sound or make the delivery of my tracks more “musical”. Would I admit to hearing a change? Absolutely, I would. I’ve been painfully honest in all of the articles that I’ve written, I made mistakes with some facts and explanations, and had to change my opinions…I love to learn new things. But I just didn’t have the same experience as my two audiophile friends.
The second thing I wanted to do was to verify that the same data was present on the output of a USB cable and the output of the Regen box. It turns out that I didn’t need to capture the output of both systems, reverse the polarity of one against the other, and see if they nulled each other out (although they did after getting the files in my Sonic Sound Blade system!). Alex Crespi, the owner of UpTone Audio wrote to me with a number of points. The second one stated, “I can tell you right now that you will not measure any difference in the bits, and if that is, as you have stated, your only criteria for judging the efficacy of the device then you are wasting you time.” OK, the processing and refinement of the signal doesn’t contribute to the DAC getting the bits more successfully.
18 thoughts on “Regen Session: Part II”
If Alex Crespi is going to admit that there is no measurable difference with or without the Regen and it’s only a subjective difference, then he is admitting it is “snake oil” but like most salesmen he believes his own B.S. Listening is a very subjective sensation. For a true statistical analysis using humans you need more than a sample size of three. But as you have stated, there is no convincing people with the facts when they have prejudged the situation. Still, keep up the good fight.
I think the reality is that the box does some processing but it’s a solution in search of a “problem”.
Digital Null Test
Mark, you can jumper the Benchmark DAC (internally) that default Digital 4 Input becomes a digital output and so can run easily a digital null test between both outputs.
I am pretty sure, that the USB Regen will pass the digital null test, but what I am saying (beginning in the former post) is, that your setup with 1 source driving 2 asynchronous DACs can’t be pass the null test. So even without the USB Regen (that should be “tested” here), the output of both DACs can’t be the same.
Even I am sure, what I mean, I would like to explain it very naively for the Joe Sixpack reader.
In an asynchronous DAC, the crystal oscillator of the USB input that creates the digital audio stream (not that of the USB protocol) does determine the pitch and speed of the track. If one DAC is for example + 10 ppm above nominal (this is still good and has nothing to do with the jitter performance of the clock, just the nominal variation), then the song is 10 ppm too fast and 10 ppm too high in pitch. This is absolutely not a problem and is not audible at all. So far, so good. If you have then a second USB DAC that runs – 10 ppm, then the song is 10 ppm to slow and 10 ppm to low in pitch. This also, is absolutely not audible and not a problem at all. But now I am coming to my point:
What will happen if you drive these both DACs out of one source and “sync” both outputs?
I leave the answer to you. Just try to do the null test (also without USB Regen) and you will see, both outputs are not the same. I am not saying, that if the sync process (for example sample rate conversion) is audible (this is a different story). All I am saying is that, you can’t compare the USB Regen in that setup, when both outputs will not have the same data (even without USB Regen).
I hope it is clear what I mean (I am not a native speaker). Good luck.
Jeurgen, I did exactly that. We changed the jumper in one of the DAC2s and I captured the digital stream with and without the Regen box. They are identical streams…and unless you believe that the same stream can produce two different sounds, there is no change in the fidelity of the output.
Thank you for your reply, but not getting to my point. I think, Russ knows, what I mean.
You are writing, that when you compared the digital out of the Benchmark DAC with the USB Regen at the input, you have had the same digital data, as without having the USB Regen at the input. Meaning the USB Regen is bit true and this is true, as every other USB hub is also bit true. But this is not the situation, how you have listened and compared the USB Regen path against the non USB Regen path. As far as I get you correctly is, that you compared both pathes only with switching the A / B box at the analog XLR out, while have both asynchronous USB Inputs at one source.
But my point is, that if you have one source and two asynchronous USB DACs connected to, the audio player has to chose one clock as the master and has to sync the second path to that clock. But for the reason, that every crystal oscillator is a bit different (even if this is only 5 ppm in nominal frequency (as mentioned above, this has nothing to do with jitter), this second signal path has to be “modified” for example in the SRC mixer in order to get both paths in sync and this is how you made your listening comparison and this is, when both digital output can’t be the same. You can confirm with asking you friend John Siau.
Juergen, I get your point. I’ll see what I can find out. Given the way clock is handled in the Benchmark DACs, I don’t imagine it will make a difference…but your point is clear.
HI, Its seems to me that what the USB Regen is doing is galvanic isolation and a bit like what some turntable power supplies do. The former decouples one circuit system from another and the latter provides a pure generated AC signal that powers the turntable motor. The galvanic isolation i have seen in USB soundcards and USB radio power supplies and antennas and they are very good in decoupling grounds, which pass currents back into sensitive circuits and more commonly in earth ground loops and hum in audio circuits, for example. The full explanation of Earth loops and galvanic isolation can be found on Wikipedia. I first came across the AC power supply for a turntable with the Linn Lingo out there are many others there are also many power supply conditioning systems around too.
This is applying analog signal thinking to a digital system…or “mixed signal system”.
So, if I was to run an MSWord text through the Regen box and print it, I could get different results or formats (same bits, sligthly different look)?
I find it hard to believe that the same bits can sound different. I mean can this box ‘clip’ the 1s and 0s ever so slightly to make them sound different? I don’t think so!
Too bad the industry keeps promoting these reptilian lubricant salespeople, it detracts from the real reason people enjoy this hobby (listening to great music and great sound whenever you please). However, and thankfully, we live in a free world with free entreprise.
Thank you for helping us novices see the real light in a consumer protection sort of way.
A few things…
You have said that Russ heard “a very subtle difference”, a “very, very subtle difference”, and that he and Chip admitted, “the change was very subtle and barely noticeable.” with the REGEN.
Here’s what Russ said, “I’m speaking for myself, but I heard a difference. With the Regen in place, I heard denser textures, clearer harmonics, and increased soundstage height and depth. The difference was subtle, but it was there.”
Russ also notes that he and Chip picked the REGEN’d system out blind. While the characterizations between what you said Russ said and what Russ said are subtle, maybe even very, very subtle, I find them very, very telling.
Here’s Russ again, “After we had listened to the system for a while I replaced the $99 LH Labs Lightspeed 1G USB cables with $5 Monoprice USB cables. Um yeah… it took all of 5 seconds to hear the difference. The level of treble glare I heard from those cheap USB cables made it sound like the volume had been turned up a notch.”
And here you are, “Russ was very disappointed in the sound of the Monoprice USB cables compared to the Light Harmonic. I heard no difference.”
Here we have a change that is in no way subtle according to Russ that you did not hear. My interpretation of your inability to hear differences other people clearly can is due to expectation bias.
Here you are again, “But OK, Michael made his point. He’s a believer…” This coming after your reference to the Machina Dynamica products. Some readers may interpret what you’ve written as I “believe” that the Machina Dynamica products (and services?) work which is simply not the case. This is also a false equivalency, comparing the Machina Dynamica “teleportation tweak” to the UpTone Audio REGEN, since the former is based on nonsense while the latter is based on solid engineering. The fact that you don’t recognize this simple fact is another reason why you and I are at odds.
What I’m pointing out here in general is that I find what you’ve written to be intentionally misleading in order to support your point of view. In terms of “believers” and the REGEN, I do not believe it works, I know it works.
Measurements. Alex from UpTone provided you with a link to John Westlake’s measurements of the REGEN where he has measured a clear and distinct difference with the REGEN. Here’s John on his measurements of the REGEN:
Now with the ReGen USB Data, notice not only the much cleaner waveform but far more importantly that there is very little “Runt” data – the background is Black. The USB Hub IC in the Regen has for the most part cleanly repackaged the data.
For sure any signal with unwanted modulation is not a good thing to have “flying around” a DAC PCB – these modulation effects will be very visible with an RF spectrum analyser looking at the Audio outputs – how the Audio system copes with this is anyone guess, and this is not the only mechanism this unwanted modulation can cross into the Analogue domain.
Note that John is talking about how the REGEN can affect the analog domain, which gets us back to that whole noise in mixed signal systems topic that I have brought up repeatedly and you ignore.
While you have more to say on this subject, if I add all of this up, I think it’s safe to say that expecting a fair and unbiased report from you on this subject appears to be wishful thinking.
But neither of the testers “guesses” were held to rigorous A-B-X conditions. I call any outcome or conclusions irrelevant. Why Mark chose to not to hold this investigation to the scientific standards he is so completely aware of I am at odds to understand.
I’m not an expert on these types of measurements Michael is referencing from John Westlake. But for me, the general concept of measuring an analog signal from the analog output of a DAC is more important for determining the total fidelity of the signal, because that’s the signal being amplified and output to the speakers. Am I correct in stating that the measurement you routinely perform in the form of signal reversal is done in the digital domain? Essentially your analysis using a step of re-digitizing the analog output? If so, then this isn’t the same signal being sent to the amplifier? If speakers and amplifiers operated in the digital domain then a final digital measurement would make sense. Maybe any issues due to noise inside a DAC in the analog domain is not a problem when re-digitizing the signal, but it might be when the analog signal is being passed to the amplifier/speaker system? I apologize in advance if anything I said was technically way off, but I can’t get past the idea that measuring the analog waveform (despite the fact that this measurement does have its own caveats) just makes the most sense to me.
Comparing signals by reversing the polarity is most always done in the digital domain. This guarantees that each sample is identical. In the analog world, the same polarity reversal can happen (we do it all the time when doing MS miking) but those are identical analog signals at the outset. The central point for me is whether the same digital data presented to a high-end DAC can possibly produce a different sounding output? For me the answer is absolutely not. Others believe otherwise.
The audible effect of the REGEN is subtle, yes, but it’s impact on the listener’s experience is profound. I just received my REGEN a couple of days ago. On lunch break from work and with the kids at school, I plugged it into my Devialet 200, started up Roon on the Mac Mini and put on a familiar song. I didn’t hear a difference at first.
A few moments later, my wife wandered in and sat down next to me. Despite the stereo being in the living room, she rarely joins me when I sit and listen to music; she was drawn in. It didn’t take too long for the beauty of the music to activate my tear ducts (also not a regular occurrence). After the song ended, she commented on the beauty of the song and we spent some time talking about the meaning of the lyrics. Something was definitely going on here!
I’ve since had a couple of more listening sessions and can state unequivocally that my system’s sound has been significantly improved by the REGEN.
On the other hand, I can imagine that if I had been present in Mark’s studio for the REGEN A/B test, I probably would have sided with Mark in not hearing a difference.
If I try to listen for what the REGEN is doing, it seems very subtle. If I sit and listen to music, the difference announces itself by relaxing my body and drawing me into the music.
I’m glad the Regen is working for you.
The USB Regen is one of the stupidest products on the planet. The guy sells specialized “musical” capacitors for god-sake. Your human test subjects are all about response bias — https://en.wikipedia.org/wiki/Response_bias (yes they are good honest guys, but we all are affected by such bias).
Run the test again using a placebo box (use the same Regen case but just insert a pass through USB cable inside), I guarantee you will get subjects hearing “differences.”
I’m willing to give the designer and UpTone Audio the benefit of the doubt that they are making a product that minimizes any fidelity loss…but if your system doesn’t have any loss in the first place, then is it a worthwhile product or not?
There’s no such thing as an audio system that has no “fidelity loss” — in reality. In theory, sure.
But if you do in fact have a system that has no fidelity loss then you would, in theory, be good to go and any change would necessarily make things worse. If I had such a system, I’d make sure to ask my unicorn to listen to it to make sure it was in fact perfect 😉