Regen: The Results Are In
It’s been an interesting Labor Day. Russ Stratton and Chip Moore from the Los Angeles and Orange County Audiophile Society were kind enough to come by and share a few hours of their holiday with me as we evaluated a borrowed REGEN box. I also heard from Alex Crespi, the owner of Uptone Audio, in a comment that you can read below yesterday’s article. He shared a numbered list of things he believes are important to know about their box and more. You should read his thoughts.
First, let me describe what we did today.
Russ brought the following equipment with him:
1 Benchmark DAC2 HGC
2 Light Harmonic 1 meter 1G USB Cables
2 Monoprice 2 meter USB Cables
1 Balanced XLR A|B Switch Box
1 Regen Box by Uptone Audio
2 Pair of 1 meter Mogami 2534 Balanced Interconnects
1 Pair of 3 meter Mogami 2534 Balanced Interconnects
Sony Vaio Laptop Computer (6 years old)
120 Gigs of SSD Storage
Daphile Player Software (LMS Music Server)
I supplied the following:
1 Benchmark DAC2 HGC
Bryston (B Power Amplifier)
Audience AU24 Speaker Cables
Mogami Balanced XLR from the A|B Switcher to the Amp
The signal path:
We hooked up the entire system without the REGEN box and listened to Laurence Juber’s “Mosaic” track (it won the 2002 “Demmy” award for “Best High-Resolution Track”), per the suggestion of a couple of readers. We carefully measured the levels with a sound pressure meter and agreed that flipping the A|B box made no difference in the sound or the level. The diagram below lays out the signal path.
Figure 1 – The signal path for evaluating the Regen box. [Click to enlarge].
The three of us then listened to the track through both of the DACs. We took turns in the sweet spot, took turns switching between the DACs, and listened intently to the sound coming from the speakers. I held back my opinion. Chip’s first response was that he, “honestly, couldn’t tell any difference.” However, Russ stated that he could. He described it as, “the Regen box version is less harsh, has better reverb tails, and is slightly more open sounding.” He admitted that the difference was very, very small and would most likely be missed if played in a system less resolving then my studio. The AIX main studio is a professionally designed and built room, tuned by industry veteran Bob Hodas, and has been called one of the best sounding rooms by SFS Media chief engineer Jack Vad.
Figure 2 – The main control room at AIX Records during the Regen evaluation.
After some additional listening, Chip also concurred with Russ that he felt the sound through the REGEN box was, “more musical” sounding. Again, he thought he heard a difference but commented that it probably wasn’t $175 worth of an improvement. “If the box is doing anything, it’s very subtle”, he said.
I heard no change at all. During one of the changes between the with and without REGEN, I thought I may have heard the space open up just a little, but I failed to notice the same effect during any of the subsequent switches…and I’m not sure whether the one with greater space involved the REGEN. Maybe my professional mastering background or my skeptical nature clouded my judgment or maybe I’m like one of UpTone audio’s customers that wrote me this morning that he returned his purchase because despite repeated tests and careful listening in his systems, he couldn’t hear any changes either.
“I received my $175 refund from UpTone Audio yesterday. I had the REGEN for less than 24 hours because it didn’t take very long to tell that it was useless. DACs tested on: Mytek Stereo 192 DSD, Teac UD-501, Musical Fidelity VLink 196, ifi nano. Headphones: Sennheiser 600. Speakers: PSB towers. Amplifier: Parasound A23. The ifi, Teac, and Mytek DACs were run directly into the Parasound A23 when listening through speakers. I’m proud to say that, according to Alex at UpTone, I’m the first and only one to return a REGEN!! I guess I just don’t know how to listen; so I hope that you vindicate my ears.” Don Walizer
We listened to some other tracks…most notably “Mujaka” from The Latin Jazz Trio, the gentlemen loved it with or without the REGEN box.
So perhaps nothing was accomplished during our unscientific test. Two audiophiles told me that heard very subtle differences and I believe them. I didn’t hear anything change despite repeated listening to a track that I’ve heard many hundreds of times. The tally for the day’s listening? Two agree that it works and one remains a doubter. What does that prove? Nothing. It’s just more information to consider when you consider purchasing an audiophile accessory. Is the REGEN “snake oil”? Certainly not to the many hundreds who have purchased one and are enjoying the enhancements they hear. As for myself, I remain highly skeptical of the whole premise behind the device and will continue to focus on making the best recordings rather then chase the illusive micro-measured audio improvements claimed by a whole variety of audio accessories.
Tomorrow, I’ll give you the skinny on whether the data I captured was the same.
30 thoughts on “Regen: The Results Are In”
Interesting results Mark. I never considered the Benchmark DAC2 HGC to be “substandard gear” but you proved that 2 out out of 3 listeners feel it is according to your beliefs.
Yes Mike, and there’s no “ID” in Freud’s psychoanalysis. Paper is paper, right ? Here’s to headlines… “Where there’s a will there’s a way”…. Marketing requires a subconscious audience and all is fluff, unless of course, you OWN it !
I also find the results to be interesting and, while having no experience with the Regen product, Chip and Russ’ comments about the Regen mirror my experience with the competing Audioquest Jitterbug. I do think there is an audible difference with the product, not jaw dropping but to my ear noticeable.
My point here is that the Jitterbug is only $50, so I can have four of them sprinkled in and amongst my various systems for about the same cost as the Regen. For that cost I don’t feel slighted at all in the cost vs. benefit argument. I am already looking forward to tomorrow’s results of the data stream analysis.
It is not clear from your diagram how you connected the USB REGEN to the input of the Benchmark DAC. Did you use the supplied solid USB A>B adaptor, or did you run another USB cable from the REGEN to the DAC? Although you may not believe in USB cables making a difference, impedance and signal integrity are very real, measurable parameters (see links in my longer reply in your article of September 5th), and the addition of a USB cable after the REGEN does reduce, but not eliminate its effectiveness.
I promise that as the months go by you will begin to see a number of similar products from other well known firms beginning to address the serious shortcomings of the USB inputs of DACs as it relates to signal integrity. There will also be a few DAC designers attempting to correct this inside their DACs, though I can assure you that almost none are doing so presently (not even your favored Benchmark DACs).
Dial the calendar back to a time when DAC designers and studio engineers dismissed jitter as concern, and then forward as a number of firms addressed the issue with reclockers of all sorts–to the cheers of many and the jeers of others. And nowadays proper clocking and low-jitter designs are the norm.
As I mentioned in the other thread, the bits are the bits, but not until after they have been decoded into such by the PHY and MAC of the DAC’s input stage. And the purpose of the REGEN is to allow that to happen in such a way that less ground-plane noise, packet-noise, and current/voltage spikes are generated inside the DAC itself as that stuff is pernicious and makes it way through to the master clock(s).
We are on the right side of history with the REGEN, and time will prove that out.
Thanks for the review.
UpTone Audio LLC
I think we did a fair and balanced evaluation today. I remain unconvinced but the two gentlemen from the society say they heard a change…albeit very subtle, so I guess I’m in the minority.
BTW The REGEN was plugged straight into the back of the DAC2 using the little adaptor…no extra cable.
If your supposition was/is that the Regen is most likely to make the biggest difference when used with the more inexpensive DACs that many of us use at our desktop, why not use one of the several inexpensive, but highly rated DACs out there in lieu of the Benchmark. And if $175 can make a $200 DAC sound almost like a $2,000 one, isn’t that worth it?
I received assurances that the Regen wouldn’t have any influence on the sound of the Benchmark. From my perspective, that was the case. The audiophiles that shared this session believed they heard “very, very subtle” differences. Adding the Regen to an inexpensive DAC will not produce the same fidelity as any high-end DAC.
So what was proven? This was not a true A-B-X test with the heard results logged and then checked later for accuracy. This is not science, I’m very disappointed Mark, you know how to hold a proper blind listening test.
We had no intention of conducting a rigorous A-B-X test. Sorry to disappoint you.
We listened carefully for a couple hours with sighted and unsighted switching between the two outputs. All three of us felt like we had enough listening time to fairly evaluate the Regen.
Any thoughts on whether the Mac/Amarra setup would have given any different results? The computer – its USB implementation and player software, etc – would seem to be variables that could be explored.
It’s not likely that using my Mac and Amarra would have made any difference with regard to the Regen. The Vaio played just fine through two USB outputs.
Quick Question. Were these tests run blind? Did the listener know which system they were listening to?
I wouldn’t say that the test was done blind. During the initial passes (all with the same tune), Russ did the switching and we listened without knowing. During the final run, we did know which was which. Hardly rigorous, I know…but I didn’t intend it to be.
The comments from the two guests sounds like the results ZI get from my Schiit Wyrd. There but very subtle. However, it was purchased to eliminate the dropouts I was having and for that, it has been very effective.
The causes of your dropouts are not likely to be traced to cable or regeneration devices…it’s about throughput.
All I know is that before the use of the Wyrd, I would have one dropout every 5 hours or so if listening. After the Wyrd (all of $99) only one every 25 hours. Worth the money IMHO. Even better, no dropouts after 100 hours with a Jitterbug!
I’m willing to bet that your dropout problems could have been solved more simply and less expensively.
Couldn’t you carry out a rigorous test on oversampling ?
I am a bit confused by the diagram. Did you use the Light Harmonic USB cables for the setup? Or did you use the Monoprice USB cables? I thought bits are bits, but are you now recommending that we all use the Light Harmonic USB cables? Or did you pick a >$1000 USB cable because you assumed most people purchasing the Uptone Regen would have a third-party USB cable? Or did you do an additional comparison between the Light Harmonic and Monoprice USB cables? I completely believe in your findings. I just wanted to understand the setup better. Thanks.
The first pass at the evaluation was done using the Light Harmonic cables. Russ had two identical cables and felt it was necessary to use them. We did change to the Monoprice cables as I was messing around with the polarity tests. Russ was very disappointed in the sound of the Monoprice USB cables compared to the Light Harmonic. I heard no difference. I’m at a loss as to how he hears something that I don’t…but that’s part of the audiophile world. For the record, I don’t own ANY high-end digital cables. I do have very nice speaker cables but use normal Mogami cable for most of my connections at the studio.
We used two identical $99 LH Labs Lightspeed 1G USB cables because 1. I had them and 2. because LH claims the cable’s bandwidth is double the 480 Mhz USB2 specification. I didn’t want cheap USB cables to be the limiting factor in our evaluation.
Have you switched only the analog balanced XLR out in the A / B switch box, or have you changed also simultaneously the USB out of you Sony Vavio between the USB left Benchmark DAC and the USB right Benchmark DAC Thanks.
The A | B switch changed the outputs of the Benchmark DACs. They didn’t change the USB cables.
The only switch we needed was for the balanced analog outputs of the two DACs. One of the cool features of LMS (Logitech Music Server) is the ability to control and sync multiple USB DACs. During setup we got both DACs operating independently then sync’d them together so that the same music file data was sent to both DACs simultaneously. There were no latency issues between the two DAC outputs that we could hear.
Thank you for clarification. I was mislead by the information, Mark has written the day before that blog, telling that he would use Amarra. Now I see, that you used Daphile / LMS Server so you can stream to different zones / outputs. Thanks.
I read about the REGEN test with interest and one thing baffles me that has nothing to do with it per se. Excellent stereo gear was used for the test so care was taken there. But the picture shows a big electronic console apparently between listeners and speakers. It seems like that could degrade the sound and stereo image. Sweet spot? How could there be one?
I suppose it could be said that the obstruction was present for both the A and B sides of the test so everything was equal. But it just seems like Audio 101 was violated.
This is my studio and is typical in its design. Professional studios have consoles in them…there is some chance for bounce but engineers sit at the right location and measurements are done from this point.
Two Master Clocks to one Source?
Hi Marc. As mentioned from you above, you have connected two Benchmark DACs with asynchronous USB inputs to your Sony Vaio computer. How do you handle, to have two individual clocks (one in each Benchmark USB Input running) that pulls the audio data out of the software player? I am pretty sure that you have good experience in studio environment, that you can have only one clock running, when connecting several digital units together and if not, you got either sample rate conversion between two different clocks or you got either buffer under-runs or over-load from time to time. Meaning, that even if you had no the USB Regen at one of the Benchmark DACs, those two streams can not be the same, because one USB Input will be the master for the audio data and the second one will not be able to control the data, so you have either not bit true (sample rate in the driver) or buffer under-runs or over-loads.
This can be solved in have for example a UPNP/DLNA server running the streams into two different “zones”, in your case, two Benchmark DACs.
So my question is, how have you came around with that? Thank you.
The setup we used locked the two DACs together in the player software. In addition, the Benchmark DACs produce their own individual clocks…they are not dependent on the incoming clock. As you correctly state, in a studio environment everything has to work from the same universal clock.
I don’t know enough about the software to answer your question. Are there any LMS experts out there who can explain how the DAC sync function works?