Dr. AIX's POSTS — 05 September 2015


A few weeks ago, Michael Lavorgna at Audiostream and I had a polite debate about the merits of certain audiophile tweaks. Specifically, we debated the whole “bits are bits” notion and the value of “regenerating” those bits as they proceed from a digital source to a high quality DAC like the Benchmark DAC2 HGC. The device in question is called the REGEN. Here’s how the popular device is described on the Uptone Audio website:

“The USB REGEN takes the digital audio stream from your computer or other music streaming device, and generates a completely new USB data signal to feed to your DAC. It accomplishes this by combining a carefully chosen USB hub chip with an ultra low-noise regulator and low-jitter clock. Importantly, it does so with ideal impedance matching—right at the input of your DAC.

If you are familiar with the variations in sound quality that come from different computer configurations, USB cables, and power supplies (no, “bits are bits” really does not apply when pursuing the audio summit), then you will immediately recognize the often dramatic effect that the REGEN can have on the connection you feel with the music.”

There is a lot more information on the Uptone Audio website. Click here to read their product description and claims of improved quality. The designer of the REGEN doesn’t believe that “bits are bits”. I do and so do many brilliant engineer/designers.

The REGEN is a very successful product for Uptone Audio. They’ve sold thousands of them at $175 each and the testimonials offered by the owners of these boxes are very enthusiastic. Can all of these customers be wrong about the perceived changes they hear? Probably not. But is the REGEN going to make a difference in a truly high-end system? No, it won’t.

The head designer of the Benchmark DAC2 HGC offered his thoiughts in an email some weeks ago. When asked whether the REGEN could accomplish what its designer claims, he wrote, “If this thing works, there is something wrong with your DAC!” I’m willing to accept that a poorly designed DAC might benefit from the REGEN, but certainly not a high-end piece like the Benchmark DAC 2.

Thanks to the kindness of one of my friends at the Los Angeles And Orange County Audiophile Society, we’re going to get the chance to check out whether the REGEN is the real deal or just another audiophile “accessory”. One of his friends is going to lend him a relatively new REGEN box for a few days. On Monday, Russ is going to bring it to the studio. He wrote, “I will bring the REGEN, Benchmark DAC2, 2 identical USB cables, XLR switch, 3 pair Mogami balanced interconnects, and laptop music server.

Least I be challenged on the procedure or technical aspects of the evaluation, I’m open to any and all suggestions about the testing. The plan is to route the signal from the music server through two identical USB cables to two identical Benchmark DAC2 units, take the outputs to the XLR switch and then to my speakers. We’ll be able to play some of my files and listen to the music via the REGEN and without it.

I also plan to capture the digital stream out of the REGEN and compare it to the same piece of music sent directly from the server to the recorder. If these two streams cancel each other out when the polarity of one is reversed, then it’s clear that the REGEN did nothing to change the bits. And if the bits are the same, then the DAC will output the same music fidelity. Game over.

I fully anticipate that I’ll confirm that “bits are bits” once again (just like the CD Illumination tweak). Although, I’m willing to accept that the REGEN and devices like it may work for some people that have substandard gear. If I find that there is no difference in the streams, then the true believers will find fault with the test, the equipment, my ears, my integrity, or the source material or something else. It happens every time.


Please visit the “Music and Audio: A User Guide To Better Sound” and submit your email to keep up to date on the launch of our my first Kickstarter campaign. I’d appreciate the support for the many months of putting together these posts. Click here to visit the site. Come back soon because I’ve been working on a new 30″ trailer video all day!

Forward this post to a friend and help us spread the word about HD-Audio Forward this post to a friend and help us spread the word about HD-Audio


About Author


Mark Waldrep, aka Dr. AIX, has been producing and engineering music for over 40 years. He learned electronics as a teenager from his HAM radio father while learning to play the guitar. Mark received the first doctorate in music composition from UCLA in 1986 for a "binaural" electronic music composition. Other advanced degrees include an MS in computer science, an MFA/MA in music, BM in music and a BA in art. As an engineer and producer, Mark has worked on projects for the Rolling Stones, 311, Tool, KISS, Blink 182, Blues Traveler, Britney Spears, the San Francisco Symphony, The Dover Quartet, Willie Nelson, Paul Williams, The Allman Brothers, Bad Company and many more. Dr. Waldrep has been an innovator when it comes to multimedia and music. He created the first enhanced CDs in the 90s, the first DVD-Videos released in the U.S., the first web-connected DVD, the first DVD-Audio title, the first music Blu-ray disc and the first 3D Music Album. Additionally, he launched the first High Definition Music Download site in 2007 called iTrax.com. A frequency speaker at audio events, author of numerous articles, Dr. Waldrep is currently writing a book on the production and reproduction of high-end music called, "High-End Audio: A Practical Guide to Production and Playback". The book should be completed in the fall of 2013.

(61) Readers Comments

  1. If computer USB output device makes no difference (bits are bits), why does Benchmark implement and advertise “UltraLock2”?

    “DAC2 converters deliver outstanding musical detail and precise stereo imaging. They employ an advanced high-headroom digital filter design, and a new high-sample-rate Asynchronous USB Audio interface. All inputs are fully isolated from interface jitter by Benchmark’s new UltraLock2™ jitter attenuation system.”
    I believe that Benchmark implementation has merit. If it does, is it not possible that other methods of providing a clean, jitter-free (well clocked) data stream have benefit?

    Apparently, BenchMark must believe the DAC’s jitter-reduction was insufficient.

    – Rich

    • Rich, I don’t believe I said that a computer’s USB output can’t have clocking or data issues. The copy that you quote from Benchmark is exactly why a box like the REGEN is unnecessary. The issue of jitter is dealt with exclusively at the DAC2 using the “UltraLock2”, which ignores the clock delivered on the incoming USB line (thus any anti-jitter efforts that are done by the REGEN or jitterbug are useless). The Benchmark procedure or something similar is used by most high-end DAC makers. John from Benchmark knows that accurate clocking is critical and that’s why they go to such lengths to do it right…without external boxes. They regenerate there own highly accurate clock.

      • You miss the point Mark. Rich makes the case that the Benchmark is too good to use in the test. You need to find a DAC that is of poor enough design that it will show the regen to be of value. Any result of your test that shows the regen to have no positive effect has already been nullified. 🙁

        • Too good to use in the test. My plan is to validate whether the REGEN box changes any of the digital information in the USB bit stream. If it cleans up the clock or regenerates the data, then fine. But if there are no problems with the clock or data in a non-REGEN stream, then the DAC doesn’t matter either.

  2. If your hypothesis is that the Regen does not change the output of the DAC, then the appropriate measurement is of the analog output of the DAC. Verifying that there is no change in the bits by capturing the digital stream only proves the devices and broken. It won’t prove the device does nothing.

    • My hypothesis is that the Regen does not change the signal being input to the DAC. I haven’t yet had someone say that identical bitstreams going into a high quality DAC will sound different. I’ve used the reversal of the polarity a number of times to show that “bits are bits”, I believe this will be one more instance.

      • So, perhaps you should be measuring the electrical signal out of the REGEN versus the signal out of the computer without it. Isn’t the contention of this company, as well as others making similar products and an entire sub community of audiophiles, precisely that the same bits will sound different going into a DAC with any of these devices in the signal path? Your polarity reversal test proves that the original data can be recovered. It doesn’t prove that “bits are bits” is all there is when the digital stream is being converted to an analog signal.

        I share your skepticism that the REGEN will have a audible effect on the output of a well designed DAC. But, as you pointed out above, that’s because the DAC designer should have already addressed the same issues. That the REGEN is likely redundant does not mean that the device, itself, does nothing.

        • If the electrical signal has low levels of noise or higher levels of noise, but I can still retrieve the digital information necessary to reproduce the analog audio, then the noise level shouldn’t matter. That is the essence of digital audio. Right? I think you may be right that devices like this will “fix” problems with DACs that haven’t been designed properly.

      • That hypothesis seems inconsistent with the claim that Regen could help a poor-quality DAC. The claim seems to be rather that it is pointless to do anything the Regen can do – if you have a high-quality DAC. And, to put it as clearly as my understanding will allow, there is more to the feed to the DAC than the ‘bits’ – they also have to be packaged, or presented, in a way the DAC can accept. Some sites then seem to extrapolate from this that a DAC can be so busy correcting ‘errors’ that it doesn’t have time or energy to produce the best analogue signal!

        Not at all a fan of the chain of causation claimed there, but it does seem that the ‘off-the-cuff’ claim that if you can perceive/measure a difference, there is something wrong with your DAC, suggests – if not requires – that the Regen is doing something – and something that can perceived as an improvement. So it then comes into the frame as an alternative to upgrading your DAC? There are real problems with assuming that everyone has one of the DACs that fit the ‘high-end’ specs. So, many of the claimed benefits could be just that – for the people who are using the Regen into lesser gear?

  3. Since your mind already seems made up that the REGEN doesn’t do anything and that you’ll find that “bits are bits”, you probably shouldn’t be doing this test. However, since you already have it lined up, at the very least, do the listening tests before you do measurements.

    Question: does your friend who owns the REGEN hear a difference on his DAC?

    • By stating what I believe in advance of the test doesn’t mean that I won’t give the evaluation a full and fair shake. Russ and a couple of other audiophiles will be hear. If they hear a difference between identical systems with the REGEN and the only difference, then I may have to acknowledge that there is something going on. However, given the information I have from the designer of the DAC that we will be using, I’m not expecting the REGEN to make any difference. If the data entering the Benchmark DAC2 is identical (and that can be tested), then my position is that the “bits are bits” argument holds. It will be fun. I have no agenda here…I’m simply stating what I know from a technical point of view. If we get into the subjective realm…I’m going to back away.

      As to your question, I don’t know…the owner is a friend of a friend.

  4. I picked up the Regen USB on a lark, to use at my desktop computer. I have not tried it feeding my Grace 902 DAC into a home stereo system ( Ayre, Parasound, Revel).

    But I gotta say, it was not a subtle difference from my Mac to the Regen and then to a GeekOut 720 and into Adam A3X desktop speakers. The first big change was in the bass reproduction—much lower, tuneful bass. And the rest of the spectrum seemed lifted a bit, clearer, more definition.

    I don’t have a clue as to how it works technically, only that to these old and skeptical ears, it seemed to be quite a positive difference in audio quality.

    • It’s beginning to look as if the Regen may make a difference when used in a low end DAC such as the GeekOut. As audiophiles, the DAC is one of the most important components in your system. I don’t consider the Geek line to be in the same league as the Benchmark.

  5. Bits are bits but are Jittery bits are not the same as non-jittery bits?

    It may well be that the UltraLock2 does make the DAC2 immune to jitter. However, DAC devices that do not use re-clocking may benefit for jitter reduction that occurs in transmission.

    If you want to evaluate the performance of devices like REGEN, the select a DAC that does not have similar functionality built-in to the device.

    – Rich

    • Jitter is not an issue when the DAC completely ignores the incoming clock…as is the case with the Benchmark DAC2 and most high-end DACs. I would be willing to test the box on a substandard DAC but I don’t have any.

  6. Hi Mark,

    The test will, of course, be conducted ‘sine prejudicio’, (Latin, not Spanish: ‘without prejudice’)?

    May I suggest that you do a test using only the two cables and two DACs right at the outset. This would prove that, for all intents and purposes, the two set ups without the Regen are indeed ‘identical’. Then you could insert the Regen into either of the set ups safe in the knowledge that the Regen was the only variable. Of course, the cables and DACs should be ‘identical’ but manufacturing tolerances may make them less than so. Establishing they were identical up front is an essential pre-requisite, I’d say.

    Good luck, and thanks for taking the time to do this test.


    • I like the suggestion that we setup the two parallel systems and establish the baseline comparison…without the REGEN.

  7. It has come to my attention that exhaustive subjective testing elsewhere shows that it will take a minimum of 2 regens in series at the sources output and one Audioquest Jitterbug at the DACs input to get close to max improvement..
    Also shown to be of positive effect is to “plug” any unused USB port with a regen, and also possibly put one on any other used USB line feeds such as printer, mouse, keyboard, etc. Effect seems related to jitter and noise reflecting between ports and degenerating the sound on the port supplying the DAC.

    • Thanks Sal…I’m not sure you’re being helpful.

  8. I think unless you use a DAC that runs on USB power you won’t notice much difference. It’s replacing the USB power with a clean power source that does the most good. Even with one of these products, how well it works will also depend on how bad the USB power is from the PC. The USB 2.0 (and probably 3.0) industry specifications are for DC voltage and its variation (5.0 ± 0.5 V) and maximum DC current (500 mA) only, and nothing related to noise, response to load changes, and the other things that specify a quality power supply.

    I have a Schiit Wyrd that competes with REGEN and it works well for that purpose, but I doubt it is doing anything to clean up the bits. It may open the data eye more to lower the BER if it weren’t zero at the DAC already, but that just means the DAC receiver wasn’t that great. You may find that the AudioQuest JitterBug will work just as well for a USB-powered DAC. All it appears to do is add bulk capacitance to the USB power from the PC to clean it up and it’s much cheaper than REGEN or Wyrd.

    By the way, I just want to be clear that when I write “DAC” I really mean the DAC system, or what’s in the box enclosure. I think you also are stating the same thing. The actual DAC IC in the box uses I2S digital data and not USB, S/PDIF, or TOSLINK directly. There are ICs that convert from those formats to I2S. How well these work will depend on how good the power source is.

    • USB powered DACs might be the problem. But then we’re back to the analog side of the equation. I think you may be right. Good points.

  9. Mark,

    I am a very strong advocate of the “bits is bits” philosophy but I have to tell you that I have purchased and installed an Audioquest Jitterbug into my system and found that it made an audible difference. Before you dismiss my DAC as having an issue, know that I have tried this in three separate systems with three different and highly rated DACs and my results are the same. Each and every time the music sounded more relaxed and pleasing to my ear.

    I cannot explain this difference and I am sure it cannot be measured, I would not discount a product such as this simply because the bits cancel everything out. John Atkinson himself when reviewing this product said that he found an audible difference as well, and could also not measure it in any way. I know what my ears tell me, and while the differences are not jaw dropping, for the $50 the product cost I do not in any way feel that I was taken advantage of and that no difference was gained. The science of this stuff is not my vocation or profession so I do not consider myself an expert on these matters, I just know what I hear and like what I hear with this system in place and I would argue that it does not prove conclusively that these products do not do anything, just because the bits cancel each other out or cannot be measured.

    • I believe you. But I must say that I’m unwilling to leave it to the “digital gods” to establish whether the sound is better or not. I’m glad it works for you and John Atkinson. However, if the data coming out of the box is identical to the data coming down a wire without the box, I’m satisfied that the box does nothing to contribute to better sound. Others may disagree…and that’s OK.

      • I have noted that with the Jitterbug product the manufacturer indicates that for it to work as they say it does, that no USB device even needs to be plugged into it. They have indicated that to have it plugged in to a USB port in your car, or in your NAS, is sufficient as the device is designed to act as some kind of antenna to filter out unwanted noise from the background of the signal, not necessarily regenerate the signal itself (at least as far as I understood the material).

        I did note an audible difference to the sound in my car and over my network streaming from my NAS, and did only have the device plugged into an unused port on the NAS. They claim a similar benefit if you plug into an unused USB port on your router as well, but I have not tried that. I have long been an advocate of filtering as I have believed to have noticed a difference there with a good quality audio/video based surge protector for instance (but I have never invested in any “power regeneration” product). What is your opinion of these types of devices, for filtering, and their effectiveness, as compared to the claims of regeneration?

        • Larry, thanks for the additional information. This is starting to sound a little like the Machina Dynamics guy, Geoff Kait and his various tweaks. If a system is properly setup and the equipment is of sufficient quality, there should be any need for filtering or power filtering. If you hear a benefit and it only costs $50, then you’ve made an improvement to your system. What you and others should realize is that high-end equipment (with properly designed power supplies) takes care of all of these things as part of their design…that’s part of what makes them high-end.

  10. Hi Mark,
    I understand your position on “bits are bits”, but I think it would be also significant if you could address the parameters of the digital signal that the REGEN is claiming to directly affected. From their website it seems the majority of the functions of REGEN are about the lowering the noise introduced by the USB bus. So I think in addition to the final waveform, you should give some thought to the the underlying noise. Could a setup like the one used by this evaluator (Archimago’s Musings) of optical cables work? http://archimago.blogspot.com/2015/05/measurements-corning-usb-3-optical.html

    Even if you think that the noise is irrelevant as long as the final bits are identical +/- REGEN, there are certainly experts on the other side that feel noise in the signal plays a role in recreating the analog waveform (even if the final bits end up being identical). I am not convinced of this argument yet, but let’s say I do believe that the noise is more important than previously recognized. The issue for me is still under what conditions is the noise high enough that a DAC wouldn’t be able to “clean up” the signal and the audio becomes degraded in some aspect. You point out the reviews and customers that find the REGEN makes a difference in their subjective experience, even a few in higher quality DACs. But I wasn’t able to find any objective noise and jitter measurements like the ones performed in the link above, pre- and post-DAC +/- REGEN. So I’m not convinced this removal of noise is useful yet. If anyone has found them, I’d be curious to know.

    So back to your DAC2, I guess the specific question from this perspective is, what if he advanced processing in the Benchmark DAC2 was able to achieve the same final noise level (and safe to assume assuming identical waveform) independent of whether the signal was passed through the REGEN? Then from this alternate point of view, we could strongly conclude that at least for the Benchmark DAC2 the REGEN would be unnecessary (unless it’s ignoring some other important parameters like impedance matching…but not sure how that could be tested). Anyway, in sum, packet noise and jitter reduction seems to be the main claim from UpTone, so that’s what I’d consider measuring too.

    • Thanks Todd. Exactly, the key question is whether the noise in a digital system can change the output analog audio.

  11. Having read all of this I’m not sure it makes sense to perform this test with only the DAC2. It’s a high end piece of gear and I’m not sure who, in their right mind, would buy it and stick an inexpensive box in the signal path for experimental purposes. I can, however, imagine someone trying it with a less expensive DAC. (Others have made this point in various ways.) I would be much more interested in hearing how this performs with a more affordable DAC.

    Maybe the question to pose to the readers is which DAC or DACS would that be?

    • Honestly, I’m personally more interested in determining whether the output of the REGEN box presents exactly the same data (forget about clock) as the straight wire. I guess using different DACs might be informative as well. Let me see what I can come up with.

  12. Hi Mark,

    You wrote: Although, I’m willing to accept that the REGEN and devices like it may work for some people that have substandard gear.

    While you have those audiophiles in for a listen, how about a two-fer and also set up a test with “substandard gear” to determine if your supposition is correct.


    • I probably shouldn’t have said “substandard” in my comment. I have an old PS Audio DAC from 25 years ago…I wonder how that would do. Although, it is hardly a “substandard” piece of equipment.

  13. Bits vs. bits: The only comparison that matters is for I2S. Do you plan to compare the I2S bit stream that is in a PC before being converted to USB against the I2S that is in the DAC box at the DAC IC? Comparing USB bits vs. bits doesn’t tell you how well the USB receiver in the DAC box is at error correcting.

    If measuring USB bits, do it at the receiver end and not at the source end if you are measuring with a high-impedance probe. USB measurements work best with test equipment made to measure USB performance, such as stuff Tektronix and Keysight (formerly Agilent and HP) make. Then you can see the eye diagram, BER, etc.

    My main point is bits vs. bits has to be measured for many formats in the data transmission steam if the BER for the I2S format is not zero, but you should start there. No errors for USB doesn’t mean no errors exist for I2S.

    • Dennis, my plan is to compare the actual audio data…not the electrical signals that are on the PC bus. I’m confident that the any reasonably high quality DAC is very good at error detecting and correcting. I will capture the output of the REGEN box in my audio DAW and the straight wire version, flip the polarity of one of them, and compare them. If they null each other, then they contained the same digital information. Game over…the REGEN box did nothing.

  14. Hi Mark,

    We, you and I, were not discussing the Benchmark DAC since I have no experience with it and the REGEN. Others do and have reported hearing improvements with the REGEN/Benchmark combo.

    As far as the notion of “substandard gear” goes, I will remind you that there are many reports from people using the REGEN with a wide range of well regarded DACs. I used the REGEN with, among others, the Auralic Vega whose measured performance is “beyond reproach” according to John Atkinson’s measurements, in what is without question a high-end system.

    I find it not only unprofessional but simply wrongheaded for you to make this claim of “substandard gear” when you admit to not knowing what you are talking about. “USB powered DACs might be the problem” being just one example from your comments here that illustrate you are unsure of the issues the REGEN addresses.

    If “…the key question is whether the noise in a digital system can change the output analog audio.”, your approach of measuring the digital data before the DAC, i.e. not measuring the analog output of the DAC, is by your own admission missing the “key question”.

    Finally, if you plan to “bow out” if “we get into the subjective realm”, this entire exercise is pointless.

    • Michael, I’m planning on doing a fair and even handed evaluation of several things when the REGEN box shows up. It sounds like you’re somewhat nervous that using a high-end DAC will blunt the effect of the unit. I trust John Siau, the designer of the Benchmark DAC2, to know how he has taken care of jitter and “mixed signal” systems to realize what I consider one of the best DACs on the planet. By “substandard” gear, I meant equipment that doesn’t deal appropriately with clock and data…and low level noise. I don’t know the designs you mentioned…perhaps they can benefit from a better clock and digital data.

      Where did I mention that I don’t know what I’m talking about? I’ve taken it upon myself to evaluate the claims made by you, the company, and others about the effectiveness of the REGEN unit. Maybe…just maybe…the REGEN box doesn’t address any issues that I have in my playback system, which uses the Benchmark.

      With your comment about the analog output of the DAC being subjected to the “noise in a digital system”, you seem to be saying that identical digital data in to the DAC might produce different sounding analog output. I don’t thing a lot of engineers would agree with that statement. Either the REGEN box changes the digital information (the digitized audio) or it doesn’t…end of debate. If it doesn’t, I’m confident that you and other will persist in claiming otherwise. But I’ll be satisfied and leave the subject.

      • “Somewhat nervous”? I can assure that is not the case. How many “high-end” DACs do I have direct experience Mark? How many “high end” DACs do you have direct experience with Mark?

        You have already made up your mind on this subject so I see no way to discuss this with you in a rational manner. Good luck with your “tests”.

        • Michael, I’ve put some effort into these tests. It’s going to occupy the better part of my Labor Day holiday. I believed, apparently mistakenly, that you would embrace my taking the time with a bunch of local audiophiles to check out the REGEN box. Given you continuing comments, I don’t think there’s any way you will be swayed. Maybe others will find the information valuable.

          • Of course I won’t be swayed Mark. You are testing the wrong thing and I already have direct experience with the REGEN.

          • I kinda thought so. On the other hand, if I hear a difference using the setup I’m going to use…I’ll be happy to admit that I’ve been wrong.

      • One more thing Mark. I am obviously, to anyone paying attention, not talking about “noise in a digital system”, as you said. I am clearly talking about noise in a mixed signal system. You know, a system with analog output. Your apparent inability to keep this distinction in mind is very telling.

        • It’s not necessary to quibble about a detail and make snide remarks. You’ve been harping on the “mixed signal system” ever since you learned that DACs (and a whole bunch of other devices) have both analog and digital signals. What I’m talking about is the digital information…the lights on vs. the lights off…stuff. If the codes coming out of the REGEN and through the USB cable without the REGEN are the same, it won’t matter what you want to call it.

          • “I’m not sure how this ‘analog’ thinking crept into the land of computer audio…” That’s what you said Mark. A DAC stands for digital to analog converter. I’ve always know this.

          • This back and forth isn’t really necessary, is it Michael. You’ve decided that you’re right and won’t be convinced no matter what…let’s move on. I’m interested in the outcome of the test and I’m sure my readers will be as well.

  15. Let’s move on? You have called the REGEN and JitterBug “snake oil” and claimed that I’m promoting “snake oil” by reporting on my direct experience with these devices. You have called into the question the integrity and/or knowledge of the people and companies behind these products — before you even had one in hand. Now, you have upped the nonsense ante by suggesting that anyone hearing a difference with the REGEN only hears a difference because they use “substandard gear”.

    Yet you are still wondering what the REGEN might do and why as is plain to anyone reading your comments here.

    You are going to perform a meaningless “test” and you think that somehow the fact that you are going to listen to the REGEN for a day or so will somehow change my opinion which is based on months of listening through a number of DACs? Wrong.

  16. Would you clearly restate your hypothesis, and how exactly you intend to prove it? I’ve seen a few variations in this thread.

    If the hypothesis is that the digital data – “the bits” – will not be changed by the REGEN, then I can almost guarantee that it will prove true, unless your sample is broken. I certainly haven’t read exhaustively on this device, but I haven’t read anyone claiming that it changes the data, and the manufacturer certainly doesn’t make that claim.

    If the hypothesis is that the REGEN does nothing to the signal entering the DAC – meaning the DAC system, not just the chip –, then you’re going to have to use some more specialized equipment, as another poster suggested. Capturing the digital data with your DAW does not answer this question.

    If your hypothesis is that the REGEN does not alter the sound out of the Benchmark DAC 2, or another DAC, then you can either rely on subjective impressions, or measure the analog output of the DAC. If we believe Benchmark’s claims for the DAC 2’s jitter immunity and careful power supply design – and I have no reason to doubt them –, then you should measure no difference.

    Your post asked for suggestions on how to improve your testing protocol. You have received many, and yet you seem unwilling to make any of these changes. If there is not enough time to revise the testing protocol, the test should be rescheduled. As with the Boston Audio Society test of “high-resolution audio,” an improperly designed test can cause more confusion than no test at all.

    • This is not about the DAC…at least to point that I’m assuming that audiophiles would be using a properly designed DAC that ignores the incoming clock (they use their own very low jitter clock) and extracts the digital data from the incoming asynchronous packets of ones and zeros. My objection to these devices is that they are not necessary in a high-end system. I want to show that the REGEN will not change the data compared to a straight wire…that’s it. If you’re prepared to “almost guarantee” that the data will remain as it was on the incoming side of the device…as we would hope…then I’ve made my point. Because if the data is the same and the clock is new in the DAC, then there is no way that the fidelity of the output will be improved. Unless you’re willing to say that the same data bits going to a DAC will produce different analog outputs, which I’m not…I’ve made my point. And nulling out the data between the two sources will show that the data is the same.

      I’ve said several times that I don’t care about the output of the Benchmark…I trust that the same program in will result in the same output.

      This evaluation is not intended to be a rigorous study or an article for the AES journal. I don’t plan to go out and acquire sophisticated testing equipment when proving the data is identical is sufficient. If you’d like to improve on the test, please do.

      • As long as the conclusion you draw from this test is only that the REGEN is not changing the data, then I don’t see why anyone should object. If you go on to write that the device does nothing, or that it can’t improve the performance of the DAC, you won’t have demonstrated that.

        • That would be your interpretation of the results. I believe that if the data is identical that it would be impossible for the sound to change…what would cause the change?

          • Electrical noise on the USB line may not be entirely isolated from the other circuits in the DAC. Whatever jitter reduction scheme is employed in any particular DAC may not be 100% effective, or in carrying out its operations on a jittery source the jitter reduction algorithm/circuit may cause timing changes or electrical fluctuations somewhere else. There may be other mechanisms I haven’t considered.

      • While I would hope that a modern high-end DAC would be sufficiently immune to jitter and noise from the USB connection affecting the analog output that there would be no audible or measurable difference from adding the REGEN, I would not be prepared to state, unequivocally, that it could have no effect until I had tested that particular assumption.

  17. I have not heard the Regen. I do have experience with the Wyrd and the Jitterbug. I bought the Wyrd quite awhile ago before all the hoopla surrounding it and the Regen in an effort to reduce the dropouts I was getting between my Mac Book Air and my DAC. This is what it is advertised to do and it worked beautifully. While I could get the occasional dropout, they were really few and far between. Several hours of listening with no dropouts. I thought I heard better sound too but would’t bet a lot on those observations. Not so with the Jitterbug. One Jitterbug on the unused USB on my Mac and the improvement was obvious. Not sure why but it works. Even better, I have had no dropouts at all in over 50 hours of listening since the use of the Jitterbug. I can live with that alone for $50. Maybe they aren’t effecting the DAC but they are sure making my Mac work better as a music server! I have heard reports of teh Regen, Jitterbug and Wyrd working with the Bricasti M1 which is a fine DAC and the Schiit Yggy which is one of the best DACs I have ever heard. The Wryd and the Jitterbug sound fantastic with my Schiit Gungnir Multibit aka the Gumby. I look forward to your listening impressions and testing Mark!

    • Joe, I’m glad it works for you…but you’ll have to send them to me and let me check them out.

    • Joe – It would be interesting to see if inserting a standard USB hub between your Mac Book Air and the DAC (without using the Wyrd or Jitterbug) reduces the dropouts.

  18. I meant with the Regen!

  19. 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, good for you. It sounds like you gave the unit a fair chance and concluded that your system didn’t benefit from an “regenerating” the stream from source to DAC. I’ll be posting my finding later.

  20. Dear Mr. Walrdrep:

    I am the owner of UpTone Audio LLC, and thus the producer of the USB REGEN (designed by my associate, the well known John Swenson). I have followed off and on your conversations with Michael Lavorgna and your postulations about what the REGEN does or does not do. Now that you are about to “test” one for yourself, it seem prudent for me to make a few comments in advance:

    1) As an industry professional with a web site and readership, it seems highly irregular and somewhat unethical for you to engage in a supposed objective review of a product (your bias and preconceived notions aside) without contacting the manufacturer (me) to ask permission or a single question about design or use of the device.

    2) 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.

    3) Despite M. Lavorgna imploring you to read the condensed “white paper” of ours (published at the front of the AudioStream review of the REGEN: http://www.audiostream.com/content/uptone-audio-usb-regen) you continue to misconstrue the mechanisms by which the REGEN works (improving signal integrity at the input of the DAC so that the DAC’s PHY chip leaves its pre-processing circuits off, thus generating less ground-plane and packet-noise and fewer current/voltage spikes on the PS).

    4) Surprisingly, you also seem to mis-understand asynchronous DAC clocking and the role (or lack thereof) played by the 24.0MHz USB clock. The clocks running the hub chip and PHY in the REGEN and the same function clock (typically 12Mhz or 24.0MHz) at the USB or Ethernet input of a DAC have absolutely no relation to the DACs master clock(s) or jitter developed inside the DAC. And NEVER do we claim the REGEN delivers a “lower jitter” signal to the DAC.

    5) If you would study the chips and input architecture common to all USB-input DACs, you would understand that what is input to these stages is not bits at all, but high-frequency voltages with variations that must be decoded. First by the PHY chip (a nasty beast with several PLLs and internal clocks at various phases), and then by a MAC USB protocol engine (in an FPGA, XMOS, Cmedia or other custom chip). And it is here where jitter and pernicious ground-plane noise starts. All the isolators and reclocking flops in the world can’t keep all of that from reaching the master clock. The place to measure jitter is at the input of the DAC chip, but I am certain you are not set up for that.

    6) If you wish to see tangible evidence of the effect the REGEN has on signal integrity, then you will have to get yourself a 3-4GHz scope and set up for eye-pattern testing. John Swenson has done just that, and we will soon publish such measurements showing the functioning. Yet just this morning someone sent me a link to post where well known engineer John Westlake (late of AudioLab) ran an eye-pattern test on a USB signal without and with a REGEN. He posted the graphs and talks a bit about what he saw:

    7) The REGEN works the same for DACs from modest to extreme. If one can hear even a slight difference with a USB cable feeding your DAC, then you should hear an even bigger difference with the REGEN–if properly used. To maintain the improved signal integrity and impedance match, the REGEN should be positioned at the input of the DAC by use of the solid USB male/male A>B adaptor provided. Putting a USB cable after the REGEN will reduce–but not eliminate–its effectiveness.

    8) While you may consider the Benchmark DACs high-end (and thank goodness they finally moved away from the ASRC of the earlier models), they are decidedly very much in the low-to-middle range of DACs our customers are successfully using the REGEN with. Some of our clients are using reference-level DACs from Berkeley, dCS, PS Audio, exaSound, Meitner, MSB, Resonessence, Auralic, and Schiit just to name a few.

    9) I agree that the bits are the bits. But they are not bits until they are turned into such–and THAT DOES NOT HAPPEN UNTIL THE OUTPUT OF THE DAC’s USB INPUT STAGE.

    So please do your readers a service and skip trying to measure data differences that I can tell you right now won’t be there. And the trouble with trying to measure the analog output is that the ADC chips of the measurement gear are going to obscure the nature of the improvement as well. That’s a whole other argument and controversy that could on ad infinitum, but even if you measure analog variations (the magnitude of what so many hear with the REGEN really ought to be measurable), how to correlate those to qualitative differences is a separate challenge.

    My telephone number is on our website and messages via the contact page come directly to my e-mail account. I trust you will show integrity in performing your evaluation and that you will lend an open ear and mind as well. The REGEN is not “snake oil.” It is solid engineering to addresses a very real issue limiting the sonic performance of the USB interface.


    Alex J. Crespi
    UpTone Audio LLC

    • Alex,

      Thanks very much for taking the time to write. I do appreciate the details you’ve provided and I’m glad that you’re finding success with your product. We’ve just finished several hours of evaluating the device in a very high-end system in my main studio (while you may believe that Benchmark is of moderate quality, I can assure you that many commercial studios use their gear…I haven’t heard better).

      I’ll be writing up the notes on what we did here today for posting. I have an open mind to things that prove their worth, have been a professional audio engineer for hundreds of commercial recordings, and have a reputation for truthfulness.

      • Hi Mark,

        From my perspective and experience, you have a reputation for Truthiness.

  21. Hello Mr. Waldrep,

    I am a customer of your productions and recipient of your Blog entries that for a time were written and emailed daily. I am also a member of Computeraudiophile.com since 2011, a beta tester on a voluntary basis for Sonic Studio based on my enthusiasm for the products since 2011. I employ a W4S Dac2 DSDse since 2011 which began as the Dac2 and has since undergone a few upgrades.

    I am also a customer of UpTone Audio having purchased not only the Regen but the JS-2 LPS and the MMK power supply and fan controller regulator. My intention in replying in response to the few articles already published about the Regen is to add to the complex of experiences proferred by users in the field. Clearly, my competencies are not in the technical realms. I am an end user who has invested in my enjoyment of music, which has always been my first priority. While I appreciate what measurements provide, my purchasing decisions are based upon sensory based feedback. And furthermore, not only what I hear, but what I feel about what I hear. You may imagine, that measurements or examination in depth of the science of sound etc. is not in my mind at the time I am evaluating what if any changes I am experiencing with the addition of a new component.

    To get to the point of my reply, after adding the complex of audio components, all together at one time and assessing what, if any, the additional components added to my enjoyment of music, the Regen as one of those components added caused a change and the effect of that change was both positive and different from what was. The difference is either positive, negative or none. In my experience overall, the difference was positive. If you were to evaluate my system and decide it was composed of certain elements and then determined as less and therefore might be susceptible to change, I can tell you that my enjoyment of the music with Amarra Symphony with iRC delivered that enjoyment. And with the addition of the Regen et al. that enjoyment was elevated a notch or two with some characteristics that needed attention as well. I respect your experience and perceptions based on your criteria. Mine are based on the enjoyment of music, and the Regen did not disappoint.


    • I’m glad that you find the addition of the Regen an upgrade to your system. From what I read on the various forums, you’re not alone. It did nothing to the sound after listening carefully for an extended period.

Leave a Reply

Your email address will not be published. Required fields are marked *