Popular Post scolley Posted June 5, 2022 Popular Post Share Posted June 5, 2022 I had an etherREGEN connected via B-side to a Mytek Brooklyn Bridge DAC for many months. The improvement the eR made to it was quite dramatic. And I saw even more improvements when I attached one on AfterDark's external OCXO clocks to the eR. But that was before I replaced my Brooklyn Bridge with a Meitner Audio MA3. Recent review here. Needless to say, the MA3 blows the Brooklyn Bridge away in terms of performance. But that's not why I'm posting. I'm posting because I was a bit surprised to find that once I put the MA3 into my system (no changes except DACs) I found that I no longer needed the etherREGEN. As far as I could tell, it was no longer making any improvements, as if everything the eR did is built into the MA3. I contacted Alex Crispi, who suggested that I give it some time to allow my ears to become more attuned to the new DAC's sound. So I listened for a few weeks with no eR, and then put it back. And I thought it might, just might, be providing a small improvement. But I was not sure. So I called a buddy of mine - the only true golden ear I've ever known, to get him to listen. This post is the result of that listening session, as documented between the equals bars (===) below. We did several tests, while we were at it. =========================================================================================== PREAMBLE There is a Closing section at the bottom of this write up. If details of the test are not required to provide confidence-in/understanding-of the tests, then just jump to the Closing at the bottom for the test conclusions. GOLDEN EAR INTRODUCTION My golden ear buddy - we'll just call him Mike - is someone I turn to when I am making changes to my audio gear, and am not sure if I'm influencing myself by expectation bias. I can never pull off a double blind A/B test with him, but I'd like to think that I've gotten pretty good with single blind tests. And often I don't tell him anything about what he's listening for, making for an even better test. And all of his passionate pursuits of home and mobile audio are limited to headphones connected to a phone with a dongle DAC. That's it. So he does not really understand what I'm doing if the test requires swapping cables around. I use him because his hearing is exquisitely acute. I attribute it to his lifelong passion for live music, which has had him attending live performances typically a few times a month, for three or four decades. When we A/B test, he typically can pick out a superior sound within a few bars of the first B track. I've seen him pick A over B (or vice versa) in as few as 3 or 4 notes. His hearing is that good. TEST BACKGROUND My previous DAC was a Mytek Brooklyn Bridge. It took its signals from Roon over a CAT6 cable. No WiFi anywhere in the audio chain. The sonic performance of the old DAC was greatly improved when I added an UpTone Audio etherREGEN network switch. Then it was further improved when I added an AfterDark external OCXO clock for the etherREGEN to use instead of its own clock. I recently purchased a Meitner Audio MA3 DAC, and was so blown away by the sonic improvements, I decided to swap the etherREGEN (and by implication, the OCXO clock) out for a well regarded - but quite cheap - SoHo 5-port router. It still sounded fantastic. I could not hear a difference, thus strongly suspected that the proprietary digital audio features (primarily jitter reducing) built into the MA3 (MFAST™, MCLK2™, MDAT2™ DSP, and MCLK™) might be doing everything the etherREGEN and clock were doing, making them redundant. So I lined up a few listening sessions for Mike, so that I could know for sure. Before the test, after having plugged the etherREGEN back in (but not the OCXO), it was suggested that I try using single mode fiber and a quality FMC as my etherREGEN's upstream connection to the network, which I did. I have lots of observations about the impacts of adding that fiber, and from switching back to the etherREGEN after a week of listening to the new MA3 DAC over the cheap router. But I'll save those for other communications, as this is dedicated to observations from Mike alone. Having disconnected my OCXO clock several weeks prior, one week before the test, I plugged the external clock into a quality power supply to warm up. Two hours before the test - without unplugging it - I moved the clock to a position where it could be used in the test. It is worth noting that after hooking the clock up to the etherREGEN, and flipping the switch on the back of the etherREGEN to make it use the external clock, and waiting a couple of minutes to let it all settle, the switch was non-functional. Apparently moving the OCXO significantly "upset" the clock, even though it never lost power. Ten minutes later though, it was functioning. Two hours after that it was used for the test. TEST PROTOCOL The tests were set up as a single blind, with Mike not knowing what we were testing other than three sets of A/B comparisons. He was told he was listening to hear which sounded better to him, A or B. I was to switch back and forth between A and B as often as he wished. An iPad with the Roon remote app running allowed him to queue all the music himself from Roon, using a Tidal playlist with music of his own choosing. It was a variety, but contained no orchestral, and little jazz. There were times in the test though, when I queued up a well recorded acoustic jazz track or two after Mike had reached a firm conclusion about that particular comparison. In each case Mike stated that the changes he had previously identified were even more evident with my selected jazz tracks. For every comparison the MA3 was used as the system's DAC. It never changed. Though Mike did not know it, we tested for three things. The sonic value of using the AfterDark OCXO clock with the etherREGEN Then, the winner of that first comparison was played both with - and without - the upstream fiber network connection (vs. copper). The copper wire connected etherREGEN alone (no external OCXO clock) was compared to an inexpensive, but well regarded, SoHo Ethernet switch. Mike could not see the changes I was making to connections, as they were all either in, or behind, the stand. THE TESTS TEST ONE: Fiber connection to etherREGEN using external OCXO compared to the same setup without the external clock. Switching from A to B was as simple as momentarily unplugging the DC power from the etherREGEN, toggling the etherREGEN's internal/external clock switch, and plugging the DC power back into the etherREGEN, then waiting 60 seconds for it to reboot and stabilize. Conclusion: Per Mike the fiber connected etherREGEN without the external clock sounded better. More richness to the sound, and the sound stage collapsed when the external OCXO clock was used. Mike had to listen to 20-30 seconds of several tracks in both A and B a few times, before he was certain. But once he knew what he was listening for, he could hear it within seconds of switching from A to B. He called it "not subtle". Remarks: IMO it has to be acknowledged that the poor performance of the etherREGEN with the external clock could be due to the clock being moved so recently before the test. This test will have to be revisited after the clock has been properly installed and and had a few weeks to settle. TEST TWO: Fiber upstream connection from the router to the etherREGEN vs. a CAT8 copper connection between those two devices. The fiber was Corning ClearCurve LC-LC duplex 9/125µm single mode fiber optic patch cable. The pair of FMC's were single-mode 1310nm SFP modules, both StarTech SFPGLCLHSMST, one SFC inserted in the SFC slot in the etherREGEN, and the other end plugged into a Small Green Computer 1 Gb FMC, which was connected to the upstream network via CAT8 Wireworld Starlight 8 patch cable. Switching from A to B was merely unplugging the switch's CAT8 upstream network connection and plugging it into the B side of the etherREGEN, and vice versa. The fiber optic patch cable was never unplugged, though it was not not functional in the B mode listening tests. Conclusion: This decision flip flopped, but wound up being certain after additional listening. Initially Mike thought the B setup (copper) sounded smoother, and that A (fiber) was a little shrill. But he was uncertain - wanted to listen further. He knew he had made a very quick observation. After further listening Mike firmly concluded that what he was hearing with A was - in fact - just better clarity - which in some passages (like trumpets) can sound shrill, but in fact sounded quite realistic. That conclusion of better realism for the A configuration was confirmed across all music played. Remarks: This lines up with the occasional on-line mention of the MA3 DAC being bright. It seems to indicate that what appears bright may just be increased musical fidelity. TEST THREE: etherREGEN connected to the network via Wireworld Starlight 8 CAT8 wire vs. a Netgear GS105 unmanaged network switch connected the same way, with both switches powered by an UpTone Audio JS-2 LPS. On the GS105 John Swenson's admonition to keep an open port between any two used ports was observed. Switching from A to B was as simple as unplugging all the CAT8 cables from one switch and plugging them into the other. Both switches remained powered up the entire time, and were at no time powered down. Conclusion: Another decision flip flop. Initial impressions were that the B setup had a better sound stage, but after additional listening concluded that the A setup was both more musical and more realistic. Remarks: Additional listening - that reversed the early preference for the B setup - was not prompted by me. It appeared that Mike was unsure of his conclusion, and wanted to hear more. But once he heard more, his preference for A became clear. CLOSING It is worth noting that all three of these comparisons took quite a few minutes listening, and multiple sets of A/B comparisons, for Mike to reach a firm conclusion. That is quite different from his typical abilities to rapidly hear differences between setups. So when he may have described some changes as "not subtle", IMO it's clear to me that to the casual listener they would be quite subtle. IMO if Mike could not hear it quickly, it was not easy to hear. Likewise in two of the comparisons Mike made some snap judgements that he eventually reversed. This - IMO - is just more evidence that to the casual listener these differences would be subtle indeed. When connected to the Meitner Audio MA3 DAC via copper Ethernet cable, the etherREGEN sounded more musical and realistic than a well regarded SoHo unmanaged switch, when both were powered the same quality LPS. Further improvements in clarity were possible when the etherREGEN's upstream Cat8 network connection was replaced with a quality single mode fiber connection. And finally, adding an AfterDark OCXO external clock to the setup reduced the audio quality, collapsing the sound stage. That final observation may not be accurate due to handling of the clock so soon before testing, and will require additional testing after the clock has had much longer time to settle. The subtlety of these differences when playing Roon streamed music through the Meitner MA3 DAC vs. the immediately audible sonic improvements that I heard when I first added an etherREGEN to this same setup (except that it was using a Mytek Brooklyn Bridge as the DAC) appears to imply two things. First, that the MA3 is bringing many sonic benefits to my system's sound that previously had required an etherREGEN to achieve. Second, that the etherREGEN - and also a fiber upstream connection - both still impart additional audible benefits when an MA3 is the DAC - vs. a Brooklyn Bridge - but the improvements are far more subtle. =========================================================================================== I'll re-do the AfterDark OCXO test soon. It's been plugged in an settling for weeks. Sorry for the long post. But I hope it was of interest. audiobomber, AfterDark., Superdad and 1 other 2 2 Digital Sources: Meitner Audio MA3 DAC, AudioQuest Dragonfly Cobalt, Roon ROCK (NUC8i5, Akasa Plato 8x case, 8GB RAM, 256 GB SSD, 1TB SSD), UpTone Audio EtherREGEN, Tidal, Qobuz. Preamplifier: none. Power Amplifier: Bel Canto e.One REF500S. Loudspeakers: GoldenEar One.R's with Herbie’s Threaded Stud Glider footers, Focal Stellia headphones. Cables: digital - Wireworld Starlight 8 Ethernet, StarTech SFPGLCLHSMST single-mode 1310nm SFP module and Small Green Computer 1 GB FMC connected by Corning LC-LC single-mode 9/125um duplex fiber; speaker - Silversmith Audio Fidelium; interconnect - Silversmith Audio Fidelium XLR; AC - Wireworld Silver Electra 7 and Electra 7; external clock - Auralis Audio Duelund Pure Silver BNC. Accessories: Power supplies - UpTone Audio JS-2’s (no stock PS’s); OCXO clock for ER - Project Clay X Geismann OCXO 10MHz Emperor Signature edition 75 Ohm; cable risers - AudioQuest Fog Lifters; power conditioning - PS Audio Stellar Power Plant 3, AudioQuest Jitterbugs; AC receptacle - PS Audio Powerport Classic, Block Audio C-Lock Lite; vibration isolation - IsoAccoustic Orea Graphite footers (amps), Symposium Accoustics RollerBlock Jr's w/Tungsten balls for DAC. Room: 26' 2" W x 11' 6" D x 7' 9" H, heavily absorbent furnishings, plaster walls, suspended and carpeted wood floor. Link to comment
Superdad Posted June 5, 2022 Share Posted June 5, 2022 Thanks for posting your wonderful report Steven! This hobby is so much isn't it? Wishing you good health and great music (especially with your new Meitner MA3...), --Alex C. scolley 1 UpTone Audio LLC Link to comment
skatbelt Posted June 5, 2022 Share Posted June 5, 2022 My advice, if it's in 'it might, just might' regions for you, sit back and don't bother... Streamer dCS Network Bridge DAC Chord DAVE Amplifier / DRC Lyngdorf TDAI-3400 Speakers Lindemann BL-10 | JL audio E-sub e110 Head-fi and reference Bakoon HPA-21 | Audeze LCD-3 (f) Power and isolation Dedicated power line | Xentek extreme isolation transformer (1KVA, balanced) | Uptone Audio EtherREGEN + Ferrum Hypsos | Sonore OpticalModule + Uptone Audio UltraCap LPS-1.2 | Jensen CI-1RR Cables Jorma Digital XLR (digital), Grimm Audio SQM RCA (analog), Kimber 8TC + WBT (speakers), custom star-quad with Oyaide connectors (AC), Ferrum (DC) and Ghent (ethernet) Software dCS Mosaic | Tidal | Qobuz Link to comment
scolley Posted June 5, 2022 Author Share Posted June 5, 2022 The point of the exercise - which I should have clarified - was to discern the degree to which Meitner Audio’s design efforts negate all the trouble we go to to clean up our digital signals. CLEARLY the benefits of other clean-up efforts remain (EtherREGEN, fiber, OCXO clocks for eR), but their value is diminished as the MA3 is bringing so much “clean up” value to the game. If every DAC could do this, there’d be no pressing need for all the products we employ to clean up our digital signals. skatbelt 1 Digital Sources: Meitner Audio MA3 DAC, AudioQuest Dragonfly Cobalt, Roon ROCK (NUC8i5, Akasa Plato 8x case, 8GB RAM, 256 GB SSD, 1TB SSD), UpTone Audio EtherREGEN, Tidal, Qobuz. Preamplifier: none. Power Amplifier: Bel Canto e.One REF500S. Loudspeakers: GoldenEar One.R's with Herbie’s Threaded Stud Glider footers, Focal Stellia headphones. Cables: digital - Wireworld Starlight 8 Ethernet, StarTech SFPGLCLHSMST single-mode 1310nm SFP module and Small Green Computer 1 GB FMC connected by Corning LC-LC single-mode 9/125um duplex fiber; speaker - Silversmith Audio Fidelium; interconnect - Silversmith Audio Fidelium XLR; AC - Wireworld Silver Electra 7 and Electra 7; external clock - Auralis Audio Duelund Pure Silver BNC. Accessories: Power supplies - UpTone Audio JS-2’s (no stock PS’s); OCXO clock for ER - Project Clay X Geismann OCXO 10MHz Emperor Signature edition 75 Ohm; cable risers - AudioQuest Fog Lifters; power conditioning - PS Audio Stellar Power Plant 3, AudioQuest Jitterbugs; AC receptacle - PS Audio Powerport Classic, Block Audio C-Lock Lite; vibration isolation - IsoAccoustic Orea Graphite footers (amps), Symposium Accoustics RollerBlock Jr's w/Tungsten balls for DAC. Room: 26' 2" W x 11' 6" D x 7' 9" H, heavily absorbent furnishings, plaster walls, suspended and carpeted wood floor. Link to comment
Encore Posted June 6, 2022 Share Posted June 6, 2022 Interesting. Usually, DAC manufacturers claim that their DACs are oblivious to the upstream chain, but they never are IME. Maybe unicorns do exist ... Have you arrived at the same conclusion as Mike, using long-term listening? EDIT: I have owned two Emm Labs/Meitner DACs, and they were anything but immune to the upstream setup. All best, Jens i5 Macbook Pro running Roon -> Uptone Etherregen -> custom-built Win10 PC serving as endpoint, with separate LPUs for mobo and a filtering digiboard (DIY) -> Audio Note DAC 5ish (a heavily modded 3.1X Bal) -> AN Kit One, heavily modded with silver wiring and Black Gates -> AN E-SPx Alnico on Townshend speaker bars. Vicoustic and GIK treatment. Link to comment
scolley Posted June 6, 2022 Author Share Posted June 6, 2022 53 minutes ago, Encore said: Have you arrived at the same conclusion as Mike, using long-term listening? Well, the jury is still out on the AfterDark clock. Haven't A/B'ed that again yet. Will report when I do. As to the other two conclusions... I scheduled the A/B'ing with Mike as soon as I determined that I could not hear the difference between the MA3 with the eR, or without it. But he was unable to come by for almost a month. In that time I did as Alex suggested and just listened to the MA3 for a few weeks, and then put the eR back. By that time my ears had grown more accustomed to its sound, and could tell that things were a bit more musical (for lack of a better word) with the eR. But not by much. In that same time-frame I installed the upstream fiber, and could tell that was better too. But again, not by a lot. So in a nutshell, it pretty much comes down to this, for me... The etherREGEN in front of my old Mytek Brooklyn Bridge made a profound - impossible to miss - improvement in sonics. But doing the same thing with the MA3, you have to listen real hard to hear a difference. So whatever the MA3 is doing, it is certainly doing something significant. Encore 1 Digital Sources: Meitner Audio MA3 DAC, AudioQuest Dragonfly Cobalt, Roon ROCK (NUC8i5, Akasa Plato 8x case, 8GB RAM, 256 GB SSD, 1TB SSD), UpTone Audio EtherREGEN, Tidal, Qobuz. Preamplifier: none. Power Amplifier: Bel Canto e.One REF500S. Loudspeakers: GoldenEar One.R's with Herbie’s Threaded Stud Glider footers, Focal Stellia headphones. Cables: digital - Wireworld Starlight 8 Ethernet, StarTech SFPGLCLHSMST single-mode 1310nm SFP module and Small Green Computer 1 GB FMC connected by Corning LC-LC single-mode 9/125um duplex fiber; speaker - Silversmith Audio Fidelium; interconnect - Silversmith Audio Fidelium XLR; AC - Wireworld Silver Electra 7 and Electra 7; external clock - Auralis Audio Duelund Pure Silver BNC. Accessories: Power supplies - UpTone Audio JS-2’s (no stock PS’s); OCXO clock for ER - Project Clay X Geismann OCXO 10MHz Emperor Signature edition 75 Ohm; cable risers - AudioQuest Fog Lifters; power conditioning - PS Audio Stellar Power Plant 3, AudioQuest Jitterbugs; AC receptacle - PS Audio Powerport Classic, Block Audio C-Lock Lite; vibration isolation - IsoAccoustic Orea Graphite footers (amps), Symposium Accoustics RollerBlock Jr's w/Tungsten balls for DAC. Room: 26' 2" W x 11' 6" D x 7' 9" H, heavily absorbent furnishings, plaster walls, suspended and carpeted wood floor. Link to comment
FrankMA Posted June 7, 2022 Share Posted June 7, 2022 I just recently purchased the Meitner MA3 as well. It is an outstanding DAC. One thing I did notice in limited testing is that the sonic signature of my direct chain was not that much different than my normal chain (which includes quite a bit of additional equipment): Direct: Unifi Dream Machine>Netgear FS105>BJ CAT6>MA3 Normal: Unifi Dream Machine>Netgear FS105>Sonore OM(Farad)>Fiber>ER(Farad)/AD(Farad)>Sablon>MA3 Whend adding the ER and/or ER/AD Clock with my prior DAC (PS Audio Directstream Sr) the difference was night & day. An obvious, immediate improvement. Not so much now. Which leads me to believe the proprietary tech in the Meitner is having an effect on my network chain (most likely jitter reduction which is really what the ER addresses according to those who know) So the question remains: has adding the Meitner closed the gap enough that I can sell off the ER/AD Clock. TBD Encore 1 MAIN: Sonore Optical Module Deluxe/Farad Super 3 (7v) > Uptone EtherRegen/Farad Super 3(9v)/AfterDark Emperor Signature/Farad Super 3 (12v) > Innuos Zenith MKII SE (ER A) > ER (B) > Meitner MA-3 DAC > Ayre K-5xe MP Preamplifier > Ayre VX-5 Twenty Amplifier > Vandersteen Quatro Woods POWER: AQ Niagara 5000 > AQ (Hurricane)/Triode Wire Labs/Shunyata Python/Nordost Blue Heaven/Audience AU24 SE/Audio Sensibilities Sig Silver CABLING: Silversmith Audio Fidelium Speaker/Sablon Pantela Reserva 2020 Lan/Ghent JSSG360 Cat 6/AQ Earth & Pegasus/Aurelis Deuland 75ohm ISOLATION: Stillpoints/Symposium/IsoAcoustics/Herbie's/PS Audio Powerbase Link to comment
Popular Post Encore Posted June 7, 2022 Popular Post Share Posted June 7, 2022 6 hours ago, FrankMA said: I just recently purchased the Meitner MA3 as well. It is an outstanding DAC. One thing I did notice in limited testing is that the sonic signature of my direct chain was not that much different than my normal chain (which includes quite a bit of additional equipment): Direct: Unifi Dream Machine>Netgear FS105>BJ CAT6>MA3 Normal: Unifi Dream Machine>Netgear FS105>Sonore OM(Farad)>Fiber>ER(Farad)/AD(Farad)>Sablon>MA3 Whend adding the ER and/or ER/AD Clock with my prior DAC (PS Audio Directstream Sr) the difference was night & day. An obvious, immediate improvement. Not so much now. Which leads me to believe the proprietary tech in the Meitner is having an effect on my network chain (most likely jitter reduction which is really what the ER addresses according to those who know) So the question remains: has adding the Meitner closed the gap enough that I can sell off the ER/AD Clock. TBD Until now, I would have treated the claim that a DAC would be impervious to the upstream chain the same as claims of useable fusion reactors. But it does seem that EMM/Meitner has finally delivered on that. That is good news. That raises the bar for competitors. As much as I love my eR, I would rather not have to bother with it. But I also think that Uptone don't need to panic--it'll probably be very difficult for others to replicate what Meitner have done here, otherwise there would be many more DACs that could do the same. Many manufacturers, including Meitner, have been claiming it for years without delivering. FrankMA and skatbelt 2 All best, Jens i5 Macbook Pro running Roon -> Uptone Etherregen -> custom-built Win10 PC serving as endpoint, with separate LPUs for mobo and a filtering digiboard (DIY) -> Audio Note DAC 5ish (a heavily modded 3.1X Bal) -> AN Kit One, heavily modded with silver wiring and Black Gates -> AN E-SPx Alnico on Townshend speaker bars. Vicoustic and GIK treatment. Link to comment
Popular Post Superdad Posted June 7, 2022 Popular Post Share Posted June 7, 2022 1 hour ago, Encore said: Until now, I would have treated the claim that a DAC would be impervious to the upstream chain the same as claims of useable fusion reactors. But it does seem that EMM/Meitner has finally delivered on that. That is good news. That raises the bar for competitors. What I find very interesting is that the Meitner MA3 uses the same ConversDigital CDMCM-2121R mConnect Ethernet>I2S input board as everyone else (Ayre, PS Audio, many others), but they manage to isolate it or do things afterwards that lessen external impacts. skatbelt, scolley, FrankMA and 1 other 4 UpTone Audio LLC Link to comment
Popular Post FrankMA Posted June 13, 2022 Popular Post Share Posted June 13, 2022 Follow up on my testing of the paths below: Direct: Unifi Dream Machine>Netgear FS105>BJ CAT6>MA3 Normal: Unifi Dream Machine>Netgear FS105>Sonore OM(Farad)>Fiber>ER(Farad)/AD(Farad)>Sablon>MA3 Listening to Qobuz files (where the ER had the greatest impact with my prior DAC - night & day really): Normal ER/Sonore path had better separation of instruments, deeper/wider soundstage and was a bit smoother. But all of this improvement was in smallish increments. Would I spend the extra $5k to get this level of improvement? No Will I pull this equipment out now that I own it and sell it? No. I still see value in keeping the audio network isolation I have in place I think it is clear the Meitner MA3 is special on a variety of fronts. Extremely musical, balanced sound across all the spectrums. It's a DAC, Pre and a streamer and integrates perfectly with Roon. Add in the fact it seems to mitigate a whole host of network gremlins and this DAC is a winner. And certainly deserves serious consideration by anyone looking for a DAC in this price range. scolley, d_elm and Encore 3 MAIN: Sonore Optical Module Deluxe/Farad Super 3 (7v) > Uptone EtherRegen/Farad Super 3(9v)/AfterDark Emperor Signature/Farad Super 3 (12v) > Innuos Zenith MKII SE (ER A) > ER (B) > Meitner MA-3 DAC > Ayre K-5xe MP Preamplifier > Ayre VX-5 Twenty Amplifier > Vandersteen Quatro Woods POWER: AQ Niagara 5000 > AQ (Hurricane)/Triode Wire Labs/Shunyata Python/Nordost Blue Heaven/Audience AU24 SE/Audio Sensibilities Sig Silver CABLING: Silversmith Audio Fidelium Speaker/Sablon Pantela Reserva 2020 Lan/Ghent JSSG360 Cat 6/AQ Earth & Pegasus/Aurelis Deuland 75ohm ISOLATION: Stillpoints/Symposium/IsoAcoustics/Herbie's/PS Audio Powerbase Link to comment
Popular Post scolley Posted June 25, 2022 Author Popular Post Share Posted June 25, 2022 SECOND - FOLLOW UP - TEST My first critical listening session with my golden ear buddy Mike had one test that did not turn out as expected. When it became clear that the UpTone Audio EtherREGEN did improved an MS3's sonics - to some extent - it would have been natural to assume that the EtherREGEN with an external OCXO clock attached (ER using the external clock over its own) would have only made that better. But it did not; it did the opposite. I had to acknowledge that though I had kept the clock plugged in continuously, I had moved it a few hours before the test. That could have effected clock performance. Personally, I doubted it, but I still felt compelled to test that again. So roughly six weeks ago I moved the clock back into a permanent and stable location, and let it settle in. Then yesterday I had Mike my golden ear buddy back for a long listening session. The results were unambiguous, heard not just by Mike, but even sitting way out of the sweet spot, were also clear to me. The results? On the best recordings, a quality external OCXO clock - that has had weeks to settle - when connected to an EtherREGEN improves the sonics of an MA3. According to Mike, "Cleaner, tighter music, Not messy or muffled. Better separation of instruments. It made me want to hear more of the music. Though it does not matter with every track. But when it does make a difference, it makes a huge difference." I would agree with that 100%. Though I do have to qualify that conclusion, as the test setup from yesterday was not identical to our test setup from my first post above. Everything was the same except that a week or so before this second listening session I inserted three Symposium Rollerblock Jr's under the MA3 (supporting the unit, lifting the stock feet off the shelf), and later swapped the Rollerblock Jr's stock balls with Symposium's Grade 10 Tungsten Carbide balls. When made the Rollerblocks were added, the sonic improvement was immediately audible. And then when I upgraded them with the Tungsten balls, that too was an immediately audible improvement. So, would the addition of the Rollerblocks influenced either of these two tests? IMO - maybe. The Rollerblock addition, and then the Tungsten ball addition, both increased the clarity and resolution - particularly imaging - of the MA3's output. Maybe if that additional clarity had been present in the first test, it might have altered the initial conclusion that the OCXO attached to the eR did not improve the output of the MA3. We can't know. So I'll modify my above conclusion above to the following. On the best recordings, a quality external OCXO clock - that has had weeks to settle - when connected to an EtherREGEN improves the sonics from an MA3 resting on three Symposium Rollerblock Jr's with Grade 10 Tungsten Carbide balls. PS - In the interest of full-disclosure, I have to acknowledge that for the first tests in my OP, I was running one version of Roon on my Roon core, and on the second test yesterday that Roon core software was a subsequent version. I honestly don't think it made a whit of difference to sonics, but that's for the reader to decide. Superdad and FrankMA 2 Digital Sources: Meitner Audio MA3 DAC, AudioQuest Dragonfly Cobalt, Roon ROCK (NUC8i5, Akasa Plato 8x case, 8GB RAM, 256 GB SSD, 1TB SSD), UpTone Audio EtherREGEN, Tidal, Qobuz. Preamplifier: none. Power Amplifier: Bel Canto e.One REF500S. Loudspeakers: GoldenEar One.R's with Herbie’s Threaded Stud Glider footers, Focal Stellia headphones. Cables: digital - Wireworld Starlight 8 Ethernet, StarTech SFPGLCLHSMST single-mode 1310nm SFP module and Small Green Computer 1 GB FMC connected by Corning LC-LC single-mode 9/125um duplex fiber; speaker - Silversmith Audio Fidelium; interconnect - Silversmith Audio Fidelium XLR; AC - Wireworld Silver Electra 7 and Electra 7; external clock - Auralis Audio Duelund Pure Silver BNC. Accessories: Power supplies - UpTone Audio JS-2’s (no stock PS’s); OCXO clock for ER - Project Clay X Geismann OCXO 10MHz Emperor Signature edition 75 Ohm; cable risers - AudioQuest Fog Lifters; power conditioning - PS Audio Stellar Power Plant 3, AudioQuest Jitterbugs; AC receptacle - PS Audio Powerport Classic, Block Audio C-Lock Lite; vibration isolation - IsoAccoustic Orea Graphite footers (amps), Symposium Accoustics RollerBlock Jr's w/Tungsten balls for DAC. Room: 26' 2" W x 11' 6" D x 7' 9" H, heavily absorbent furnishings, plaster walls, suspended and carpeted wood floor. Link to comment
Popular Post JohnSwenson Posted June 28, 2022 Popular Post Share Posted June 28, 2022 On 6/25/2022 at 9:18 AM, scolley said: I would agree with that 100%. Though I do have to qualify that conclusion, as the test setup from yesterday was not identical to our test setup from my first post above. Everything was the same except that a week or so before this second listening session I inserted three Symposium Rollerblock Jr's under the MA3 (supporting the unit, lifting the stock feet off the shelf), and later swapped the Rollerblock Jr's stock balls with Symposium's Grade 10 Tungsten Carbide balls. When made the Rollerblocks were added, the sonic improvement was immediately audible. And then when I upgraded them with the Tungsten balls, that too was an immediately audible improvement. So, would the addition of the Rollerblocks influenced either of these two tests? IMO - maybe. The Rollerblock addition, and then the Tungsten ball addition, both increased the clarity and resolution - particularly imaging - of the MA3's output. Maybe if that additional clarity had been present in the first test, it might have altered the initial conclusion that the OCXO attached to the eR did not improve the output of the MA3. We can't know. So I'll modify my above conclusion above to the following. On the best recordings, a quality external OCXO clock - that has had weeks to settle - when connected to an EtherREGEN improves the sonics from an MA3 resting on three Symposium Rollerblock Jr's with Grade 10 Tungsten Carbide balls. PS - In the interest of full-disclosure, I have to acknowledge that for the first tests in my OP, I was running one version of Roon on my Roon core, and on the second test yesterday that Roon core software was a subsequent version. I honestly don't think it made a whit of difference to sonics, but that's for the reader to decide. I can very well believe that the roller blocks can make a big difference, I've seen it in my own system and have what what I think is decent explanation: crystals are by definition sensitive to vibration, that's what they do! The "better" the crystal the more effort goes into isolating the crystal from as much external environment as possible. For example that OCXO you have almost certainly has the crystal mounted in a vacuum (air currents can really mess up a good crystal). They are mounted in very carefully designed ways trying to block out external vibrations. Unfortunately they are not perfect. They are very good at isolating vibrations above about 10 Hz., but that leaves sub 10Hz noise. This is commonly referred to as seismic noise. It doesn't just mean earthquakes, it comes from many sources : ground expanding contracting with more or less sun, a truck rolling by three streets away, someone walking on the floor of your hose etc. Why does this matter, how can you possibly hear this? Well this noise is modulating the output of the crystal, in other words adding phase noise to the clock output. Roller blocks are very good at isolating this seismic noise, one of the very few systems that can do this. That's why putting the roller blocks on the clock made a such a difference, you were now actually hearing what the clock was capable of producing when it wasn't getting modulated by the seismic noise. John S. Superdad, roman410, scolley and 1 other 2 2 Link to comment
scolley Posted June 28, 2022 Author Share Posted June 28, 2022 18 hours ago, JohnSwenson said: Why does this matter, how can you possibly hear this? Well this noise is modulating the output of the crystal, in other words adding phase noise to the clock output. Thanks a mil’ for the great explanation John! The Rollerblocks are under the MA3 DAC, and boy what a difference. Now I’m wondering if some should go under the OCXO clock connected to my eR? Digital Sources: Meitner Audio MA3 DAC, AudioQuest Dragonfly Cobalt, Roon ROCK (NUC8i5, Akasa Plato 8x case, 8GB RAM, 256 GB SSD, 1TB SSD), UpTone Audio EtherREGEN, Tidal, Qobuz. Preamplifier: none. Power Amplifier: Bel Canto e.One REF500S. Loudspeakers: GoldenEar One.R's with Herbie’s Threaded Stud Glider footers, Focal Stellia headphones. Cables: digital - Wireworld Starlight 8 Ethernet, StarTech SFPGLCLHSMST single-mode 1310nm SFP module and Small Green Computer 1 GB FMC connected by Corning LC-LC single-mode 9/125um duplex fiber; speaker - Silversmith Audio Fidelium; interconnect - Silversmith Audio Fidelium XLR; AC - Wireworld Silver Electra 7 and Electra 7; external clock - Auralis Audio Duelund Pure Silver BNC. Accessories: Power supplies - UpTone Audio JS-2’s (no stock PS’s); OCXO clock for ER - Project Clay X Geismann OCXO 10MHz Emperor Signature edition 75 Ohm; cable risers - AudioQuest Fog Lifters; power conditioning - PS Audio Stellar Power Plant 3, AudioQuest Jitterbugs; AC receptacle - PS Audio Powerport Classic, Block Audio C-Lock Lite; vibration isolation - IsoAccoustic Orea Graphite footers (amps), Symposium Accoustics RollerBlock Jr's w/Tungsten balls for DAC. Room: 26' 2" W x 11' 6" D x 7' 9" H, heavily absorbent furnishings, plaster walls, suspended and carpeted wood floor. Link to comment
Mihaylov Posted July 3, 2022 Share Posted July 3, 2022 On 6/7/2022 at 8:29 PM, Superdad said: What I find very interesting is that the Meitner MA3 uses the same ConversDigital CDMCM-2121R mConnect Ethernet>I2S input board as MytekBrooklyn Bridge DAC 😏. Moreover, it is not clear where the difference in the sound of both these devices with ER came from. The CDMCM-2121R is a self-sufficient board, an ethernet connector is directly connected to it and there are no built-in switch in Maitner. Link to comment
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now