Thank you Matt for the thoughtful update.
As I suspected, after the 400 or 450 hours mark, Aeris would start a gradual normalization of treble artifacts. I still expect some fluctuations throughout the audible bandwidth for a while. My experience with Aeris, as well as other Rowland components is that they are late bloomers... They continue to break-in and bloom at least until the 1200 hours mark. At 700 to 800 hours, they give a semi-reasonable idea of their capabilities, but they still do not yield their best.
BTW, I have confirmation from the Rowland factory that Aeris delivers its best performance through its SPDIF inputs, which are transformer-coupled for optimizing common mode noise rejection, galvanic isolation, and impedance stability... This will be particularly effective if Eris is fed from a transformer-coupled source. It is interesting to note that Rowland feeds Aeris mostly through the Bryston BDP-2 server/player, which has implemented transformer coupling on its SPDIF and XLR outputs.
No, I am not using the Bryston player in my system yet... I am a dinosaur, and am still using the Esoteric X-01 as a source.
G. |
Perhaps it is just me Grannyring... But most pieces I had had in my system took that long, or even more... Esoteric X-01 and K-01 for example, both about 1200 hours... And my Vienna Die Muzuik speakers took 1500 hours... Even my old
ARC Ref3 took about 1K hours to fully bloom. I remember evaluating a very expensive twin-box player once upon a time... I did throw in the towel when I reached the 2K hours mark and realized that the device had stabilized, with no musicality to show for the effort.
I fully admit that I am excruciatingly patient when evaluating a component... If in the end I really like a device, I then use it for several years... So, even 1500 hours, becomes a perfectly reasonable investment of my time, when I consider it against the long run enjoyment.
I suspect I do not even know how to spell cwik gruddiphicationn *grins!*
G. |
Hi again Grannyring, When I break in a component, the device is exercised 24/7 until it stabilizes... For Aeris, I fed a signal into it, such as a break-in CD, and let the DAC feed into the Criterion preamp on 0 volume.... In one week I had 168 hours already... The 1K hours mark was just shy of 5 more weeks away, and after one more week I was close to 1.2K hours... I sampled results every few days. G. |
Hey Matt, I just discovered your DAC thread... So I'll follow your exciting front-end adventure!
Saluti, Guido |
Uhrn... Shouldn't this topic of single ended to XLR on CDp deserve its own thread? G. |
Hey Mat, just in case you connect multiple DACs to your darling Rowland Criterion... XLR inputs 1 and 2 have been exercised the longest. XLR 3 is also well broken in. RCA 1 has also had good use. I have not exercised any other inputs. Saluti, G. |
Oops, you are absolutely correct Matt: the RCA input I have used is input 5. Guido |
Hi Matt, it is likely that Rowland Aeris will continue to bloom until, or possibly through, the 800 hours mark.
With any component, Until a device is fully stabilized, there is the danger of mistaking temporary early artifacts as sonic signature.
G. |
Excellent question Charles. I freely admit of being pathologically picky when it comes to break-in process. I deem a device to be broken in only when I no longer perceive any further change in sound over the span of about one week of 24/7 operation, regardless of what the manufacturer has told me, or has published.
during the last 10 years or so, I have broken in a number of components, including 4 digital front end devices.. Esoteric X-01, GamuT CD3, Esoteric K-01, and recently Rowland Aeris DAC. Aeris is the very first front end that in my experience may have stabilized before crossing the 1000 hours mark.
Admittedly, Aeris does contain a lot of technology that is well known to take time to stabilize, like input transformer an 6-layer ceramic boards, yet break-in seems to have been a little faster than other front-end devices I have had in system. If you are interested, there are 27 mostly technical entries on Aeris hardware and firmware in the Rowland Knowledge base at:
http://jeffrowlandgroup.com/kb/categories.php?categoryid=205
Salluti, Guido |
Hi Charles and JaFox, my long break-in opinions are based on a break-in tracking spreadsheet correlated with a dated break-in log that I kept during listening sessions throughout the process... I largely applied the same test track that I tend to report on for published articles. As pointed out, memory of past events by itself is unreliable.
Between the 450 and 800 hours mark, the last remenents of treble peakiness and transient stresses in Aeris disappeared, while harmonic coherence, extension, imaging/staging, and overall musicality continue to grow.
G. |
|
|
Hi Matt, at 170 hours, Aeris is likely just getting into adolescence... I suspect that you will experience some sonic cycling over the next few weeks, with some harmonic and treble anomalies showing up and fading out periodically. The good news is that performance dips will get progressively shorter and shallower each time, while magic will get correspondingly stronger and longer.
Guido |
Hi Agear, Rowland Aeris (including external power supply) lists in the US for $9800. World wide dealers are found at:
http://jeffrowlandgroup.com/us/distribution-3.html
Input protocols as follows:
1x USB — accepts up to 24 bit PCM at 44.1, 48, 88.2, and 96 kHz sample rates
1x Toslink — accepts up to 24 bit PCM from 11 - 192 kHz sample rates
2x SPDIF — accepts up to 24 bit PCM from 11 - 192 kHz sample rates
Regarding Tenor and class D.... Thankfully, there is a world of difference between the sound of raw ICEpower low-end implementations of old, and the involving musicality yielded by some of the recent implementations of NCore NC1200 like the Merrill Veritas used by Matt. I have not had the opportunity of comparing Veritas with Tenor, but had Veritas in my system for several months, resulting in my more than favorable PFO review:
http://positive-feedback.com/Issue68/merrill_audio.htm
Guido |
Thank you Agear, unfortunately I never heard a high end digital amplifier in my life, so I have no idea how one would sound like.
I have two biases.... Equipment which will not serve as a secondary furnaces... And components that conform to my ideal of music.... A few class D amplifiers do.... Veritas, some Rowland amps, the new Bel Canto Black, possibly the Mola Mola... A few solid states, like Solution....
I also enjoy the sound of a handful of valve based amps, like the VTL 450 and Siegfrieds and the ARC reference series... But am not interested in baby sitting and rolling aging tubes, hence I will not seek tubed components for my system.
G |
Thank you Steve. Yes, I have amps with ceramic boards, Jensen caps, and lots more esoteric goodies inside... Amazing sound from the get go... And then they kept improving.... By 1500 hours, my amps were awesome. The delightful Merrill Veritas took a little less: By 1K hours they were stable. This has been my consistent experience for the last 30 years.
Of course, Your OD DAC might very well have a very different break-in curve... Unfortunately, I have not had the opportunity of verifying its break-in pattern and its sound in my system. G. |
AudioLabrynth, Audiogon shows the viewcount for all threads that you have started. If you simply respond to a thread instead, its viewcount is not shown in your threads list.
Guido |
Hi Agear, a few posts ago, you brought up an interesting technical point... Electrical and mechanical grounding. I looked this up in the Rowland Knowledge base, and here is what I found in the Aeris section....
Aeris uses very small surface mount devices (SMD). According to one article, a benefit of SMD is "allowing these components to be located as close as possible to corresponding ground planes". See:
http://jeffrowlandgroup.com/kb/questions.php?questionid=622
More on Aeris grounding... "In the Aeris 6-layer PCB, two complete layers are dedicated to ground. The clock signal is isolated from the signal path by these two grounding layers. Two layers are dedicated to power distribution, and two layers are dedicated to signal distribution. As a result, the clock signal cannot be contaminated by the audio signal." ... There is more on electric grounding and 6-layer PCB design at:
http://jeffrowlandgroup.com/kb/questions.php?questionid=621
On mechanical isolation of the power supplies... "Aeris is powered by two separate low noise, passive power factor corrected, switch mode power supplies (SMPS) – one for analog and one for digital. Both are housed in a single machined external enclosure for maximum mechanical and EMI/RFI isolation." See:
http://jeffrowlandgroup.com/kb/questions.php?questionid=619
In general, Aeris circuit sections are buried inside isolated pockets carved in the aluminum ingot of the main chassis.... But this is nothing terribly new... Rowland has used monolythic chassis for mechanical grounding/isolation for a couple of decades.
Other manufacturers may have used totally different techniques... Of course, in the end, it really does not matter... What is important is whether we emotionally drawn by the sound of a component or we are not.
G. |
Hi Matt, you are right... While Criterion is far from being a "bright" pre, it does not share the slightly dark golden house sound of classic Rowland products from the 1990s and the earlier part of the 200X decade... Fact is that, with some exceptions, the "dark Rowland sound" is becoming a bit of a charming urban legend.
On the other hand, I expect that Aeris's juvenile exhuberance in the treble region should start to be reabsorbed once you pass the 450 hours mark.
Concerning Final Drive and its transformer coupling... It is worth pointing out that both Aeris and Criterion have trnsformer-coupled balanced I/O... on Aeris transformer-coupled input is the SPDIF 75 Ohm coaxial. Conversely, Veritas inputs are not transformer coupled as far as I know. G. |
|
Erik, Matt indicated that he would be away until Wednesday... Updates might need to wait for a couple more days. G. |
|
Merrill Veritas have no transformer-coupled inputs nor outputs. Rowland Criterion linestage is coupled on all balanced inputs and outputs. Aeris has transformer coupling on SPDIF inputs, and on balanced outputs. It would be indeed interesting to gage the effect of FD between Criterion and Veritas, between Aeris directly into Veritas, and between OD directly into Veritas.
G. |
Matt said... "I am pretty sure the Rowland Aeris is also software driven. I bet that the right firmware update could make the Aeris untouchable."
You are correct Matt, Rowland Aeris is driven by a large FPGA which performes a variety of master functions, including FIFO jitter elimination and control of a 24-bit DAC module.
I have no information about what a hypothetical Aeris FPGA firmware update might achieve... Meantime, the best way to use Aeris is to feed it through SPDIF... Aeris performance is optimized through its transformer-coupled SPDIF coax input, ideally fed from a transformer-coupled source, such as the Bryston BDP1 or BDP-2 music servers.
Purely conjectural on my part, sound may be further enhanced by the use of an SSD as a storage medium, to eliminate jitter caused by inter-sector disk head seek delays.
G.
Admittedly, in the end the above does not imply that you would necessarily prefer Aeris over OverDrive or other DAC... DAC preference is, very much like paraphrasing female beauty, in the... ears of the beholder *grins!*
G |
Hi Matt, very nice that your MacMini already on SSD. Friend of mine made comparison in last couple of days between hard drive (do not remember brand) and Samsung 840 EVO 1TB SSD... He asserted that the SSD delivered harmonically denser images... I believe he may be using a MacMini re-hosted to Windows 7 into a Teac UD-501 DAC via a Prana USB cable.
Here are some additional twists for the permanently obsessive.... Does the SSD technology standard matter? 2.5 inch, vs MSATA, vs M.2... What about the SSD controller? What about the SSD brand, series, model? All things that potentially we can obsess about.
Admittedly, I am still a dinosaur... I am feeding Aeris SPDIF through the Esoteric X-01 coax output.... I do not believe X-01 SPDIF to be transformer-coupled. I will probably let music server technology steep for a bit yet before I take the plunge... Amongst other things, I need to make sure that the UIs to control the servers are fully interoperable with my non-visual access technology.
G. |
Thank you for the Esoteric arcana Alex... Did not know that X-01 SPDIF out was transformer coupled... At some point in the future, I might replace X-01 with a music server.
Saluti, G. |
Matt, Audiogon is a bit of a "homegrown'" posting system... There is an ability for editing a post, kind, sort of, and it depends... That is... The thread must be on a single page, and your post to be edited must be the very last post on the thread... If your post is on page 2 or beyond, Or someone posted after you, you are out of proverbial luck.... I have communicated the problem to Audiogon more than once, and met with summary dismissal of my concerns. G. |
Hi Agear, good question... But I suspect that various Esoteric transports might yield different results into a given DAC... E.g. X-01, X-03, K-01, K-03, P-02, P-01 Grandioso have different transport subassemblies, different electronics, and different mechanical packages.
G. |
Agear, unfortunately generic questions do not often have attractive universal answers... The audio devil always lurks in the proverbial details.
Hence the endless variables... Which CDPs, which computers, which connections, which user preferences, etc...? G. |
Matt, I could not have said it better myself.... You never asserted that you were going on a quest to discover the "best" digital front end.... You stated very clearly from the very beginning that you were trying to identify the DAC component that you prefered for your own system, cabling, and musical/sonic preference... I do not understand what is so difficult about this concept folks!
Just one minor correction... The Rowland team uses Aeris with SPDIF input because they prefer its sound/music than from USB wires of the same length and brand.... The ability of running longer lengths of wiring using SPDIF 75 Ohm coax without degradation is purely icing on the audible cake.
All... Whether Empirical on SPDIF would be preferable to Aeris in a prticular system and for a particular end user, or perhaps the opposite, Is something I will not venture to guess.... I have no direct experience with OD, and any such "beliefs" have no valid foundation without live experience in a controlled environment... But all of this remains a moot point, as Matt's front end feeds a USB protocol into the DAC, and will not change in the foreseeable future.
Besides, there is a new fascinating contender in Matt's grand quest... Alex's APL DSD... Let us see how Matt likes it once it stabilizes completely.... Matt, I am very much looking forward to your well thought findings... Alex, what do you expect to be the complete break-in time for your lovely DAC?
Saluti, Guido
|
You are correct Alex, Aeris does not support bit rates greater than 96K under USB input... However, It has been reported to me that Aeris is preferable with SPDIF input over USB for material that is supported by both input protocols... On the other hand, my X-01 seems to have only SPDIF output, so I cannot verify the assertion in my own system.
I am not surprised that break-in time for your DSD might exceed 700 hours... I have observed the same with Aeris... I hope that Matt will be able to afford enough break-in time to stabilize DSD completely.... And then assess his preferences against OD, and... Aeris... Unless Aeris has been sold by then *Sighs!*.
G. |
Hi Matt, any chance of you performing break-in of OD and DSD-S in parallel? Would shorten the total process.... E.g. If your MacMini were capable of engaging two USB outs at a time.... Or you fed one device via MacMini and the other via any other source, like a cheap computer or CDp.... The outputs of Both DACs would need to be connected to an active load.... But your Criterion has plenty of spare inputs.... For the purpose of break-in, it really should not matter the quality of the connecting cables....
One of the side benefits is that that you might be able to engage the devices for more than a total of 700 hours each.... And still take less time than your estimate.
G. |
|
Hi Alex, in Aeris, jitter removal is executed in an FPGA rather than inside the DAC.
BTW, my own system has the opposite limitation of Matt's... only supported digital out from my battle-hardened X-01 seems to be SPDIF.
Cari saluti, G. |
Hi AGear, somehow the URL of the AA thread in question did not make it to your post... Mind reposting it?
Thanks, G. |
Hi Matt, I am not surprised about Veritas small current draw... Their efficiency is probably above 90%... Just to be on the safe side, have you tested current draw with some tracks that have large scale orchestra transients? E.g. 4th movement in Mahler's 1st symphony... Or those dreadful cannon shots in the Tchaikowsky 1812 Ouverture... Very loud but short duration transients might draw more current than otherwise seemingly loud sostenutos. G. |
Hi Matt, fabulous report as usual! Would you mind giving a quick test of DSD-S in current break-in state using the identical USB wires and PC as you used for the other DACs.... So to isolate the properties of the box from those imparted by the cords.
No, I have no conjecture on whether you will find its sound to be more/less/samely appealing.
G. |
Paul79, you will find Matt's system discussed in detail at:
http://forum.audiogon.com/cgi-bin/fr.pl?vaslt&1386218525&openmine&zzGuidocorona&4&5#Guidocorona
G. |
How refreshingly self-afacing BigKidz.... What is the name of your company?
G. |
Alex said... "Yes, I have read your posts that you don't like the sound of AKM DACs, but try to convince the owners of Esoteric K-01, for example, that their player
doesn't sound good because it has AKM inside."
Bingo! A couple of years ago I had an Esoteric K-01 loaner in my system for quite a spell, and its music with 4X upsampling and Slow Delay 2 (apodizing) filter was sheer magic to my ears.
Let us not get too hung up about the quasi religious propriety of using a particular chip, electronic part, sub-circuit design tactic, etc... The totality of a component is often much more than the raw sum of its parts... While sometimes, sadly enough, it ends up being much less than the aforementioned sum of such promising parts.
G. |
Steven Alex, on your respective DACs, what signal processing parameters can be controlled by user selectable settings on front panel / remote? E.g...
* Upsampling rates,
* Filter types and subtypes,
* filters on/off,
:* Other?
Saluti, Guido |
Alex, what user-controllable options are available via remote or front panel on DSD-S, besides PCM2DSD 64 vs 128 conversion?
G. |
Thank you steve, could you please give more detail about the OV digital filter selections, and preemphasis settings.
B |
|
BTW... I enjoy learning about OV from Steve... DSD-S from Alex... And K-01 from Esoteric... Please do continue to sing the refined virtues of your very own designs!
Yet, Call me oldfashioned, I am deeply uncomfortable reading engineers publicly and shamelessly peeking into each other's technical "private parts".
G. |
Hi Audiolabyrinth, Criterion used to list for $18.9K, but has completed production... The current Rowland flagship is Corus, listing for $14.9K.... The Capri S2 at $3,950 is also a wonderful linestage.... A little warmer than Corus, and supports an optional internal DAC card with SPDIF input.
G. |
Thank you Matt for your ever-thoughtful observations... Am I correct to assume that you did not go directly from DSD-S into the Veritas, but you had your Criterion line-stage in the chain instead?
Guido |
I second Steve on his recommendation for large/thick/dense area rugs.... Their beneficial effect can be further enhanced by resting them on top of one or more layers of non-skid latex sheets... Yes, I did use a latex non-skid underpad under a thick woolen rug when I had hardwood in the living room.... Rug was better than naked hardwood, and non-skid pad + rug was better than rug alone.
G. |
Matt, Here is an interesting development… Esoteric has announced/released new versions of its fabulous K-01 and K-03 players, to be called K-01X and K-03ZX respectively. Apparently they use some trickle-down technologies from the Grandioso multibox flagship players. I absolutely loved the original K-01, with slow Delay 2 filter and 4X upsampling, but I do not know how the K-01X DAC component driven by your MacMini via USB would compare to OD or even Aeris.
http://www.esoteric.jp/products/esoteric/k01x/indexe.html
Compatibility with high resolution source material is delivered through a free High Res audio player, downloadable from:
http://www.esoteric.jp/products/esoteric/hr_audio/indexe.html
Saluti, G. |
Keith, you should check details on the page that I have posted above. Seems there is only a partial upgrade path for original K-01 K-03 owners. I am not quite sure of the US MSRP of K-01X and K-03X... Probably K-01 hovering around $20K, but I am guessing.
G. |
Thank you so much Alex... Let us keep in touch! G. |
Correct on K-01 breakin.I suggest the following:
Set device to
4X upsampling,
S_dly2 filter,
digital out active,
display dim automatically after 5 or 10 seconds.
Connect K-01 via XLR outs to Criterion... Simultaneously with SPDIF to OV... The two outputs can be broken-in in simultaneously.
Give the newborn a steady diet of high-fiber material for 1K hours.... E.g. Stravinsky Rite of Spring, mixed with a bit of Purist Break-in CD, may be a touch of punk rock for good measure.
Teste taste periodically the results on the two outputs.... Note that for best results on the XLR outs, the digital outputs should be deactivated during critical listening.
After basic break-in, each further filter/codec that you wish to develop to its full potential, requires its own 500 hours. So....
500 on DSD
500 on FIR2 + 2X upsampling
I am not sure if FiR1 and S_dly1 require thir own independent breakin time.
Saluti, G.
|
Audiolabyrinth, early/intermediate impressions throughout break-in are actually interesting, as they show the evolution of a device into sonic maturity... And might just prevent audiophilic panic in impatient new owners who do not experience nirvana at first sound. G. |
Hi Keith, I do not own a K-01 at this time. My past findings are summarized in some of my previous posts. Sorry... I am not looking for "believers" of any sort... A healthy dose of Skepticism is expected and welcome. If anyone has experienced stabilization of K-01 in significantly less than 1K hours, please post your detailed findings here or on one of the various K-01 related threads.
G. |
