Recommend a high-end DAC and amp stack to a genuinely curious objectivist

[MrHaelscheir]

tl;dr: Recommend an alternative to the Holo Audio May KTE and Bliss KTE stack that would be most likely to "convince" an "objectivist" upgrading from the FiiO K9 Pro ESS with binaural HRTF DSP of "R2R and class A goodness" or other subjective claims of improvement, particularly "huge soundstage", "holography", "realism", "dynamics", and "resolution".

This post is structured with context regarding what kind of a listener I am, my experiences, my expectations, what I have tried, and then the details of my inquiry.

If you happened to have been following my posts on Head-fi, you would see that I am an "objectivist" of the stance that any subjective differences not able to be revealed by an electrical or acoustic "null test" are most likely to be psychological in origin, which is not a "bad thing" so long as one does not make the jump from "subjectivism" into making objective claims about the electrical or acoustic performance of certain audio gear and recommending similar results for others.

https://www.head-fi.org/members/mrhaelscheir.572134/#about links posts highlighting milestones in my audio journey starting with the HiFiMan Arya Stealth, Meze Elite, EQing by ear, crossfeed, in-ear mic headphone HRTF and acoustic cable measurements, HE1000se diminishing returns, and then refinement of my binaural head-tracking. https://www.head-fi.org/threads/rec...-virtualization.890719/page-121#post-18027627 (post #1,812) covers my latest developments in what is effectively like a DIY Smyth Realiser A16. This all has been in service of reproducing the live experience of symphonic music through headphones, or as some describe it, the "you are there" experience. I regularly attend live classical concerts to check against my binaural head-tracking setup and EQ and know that the target level of "resolution" or "soundstage" is certainly not "infinite" unless regarding an asymptote approaching the music simply sounding real. The imaging with personalized binaural head-tracking is simply "unreal", though I know there are still flaws with my HRTF measurements and the binaural rendering software including limitations to the sense of distance, whereby though some live performances can be uncannily tonally similar to my memory of the recording, I know that most recordings still bear that sense of "recordedness" (not enough vividness or air and separation) sitting between myself and full immersion. I do have a pair of Genelec 8341As, but having a proper listening room with acoustic treatment is a long way away (without such, there are indeed some imaging and tonal quirks and some detracting to clarity), and I will likely never have a sufficiently large system to convincingly image an orchestra (when you don't want to wait months to relive a particularly moving concert hall experience), though I may be able to take HRTF measurements from 16 meters away and see if that helps image things further away. Basically, as far as I have come, I am still "discontent", which goes back to when I felt that my EQed Arya Stealth still didn't "sound real enough" (though some recordings did give me an "I am on the stage" experience, though maybe only once before the hype wore off), and discovered what I did when I auditioned the Meze Elite et al.

• So far, upgrading the headphone has merely given me measurably improved distortion performance as with the EQed Meze Elite (which in my latest findings easily rivals the DCA Expanse and Stealth, possibly outdoing them for multitone distortion), and improved comfort, there being no major "resolution" improvements that could not be emulated by matching frequency responses with in-ear mics (EQ based on measurement rigs is not guaranteed to match the frequency responses for your own ears, mind it was such an error that made the Meze Elite sound so "special" and engagingly "sweet" (600 Hz elevation) when I auditioned it).
• I am the sort of person for which the Sennheiser HE-1 in the 10 minutes I got to listen to it (after spending too much time first doing an in-ear measurements at least for valuable and impressive frequency response results) did not impress me in presentation or vividness compared to the sound of my EQed Meze Elite with FiiO K9 Pro ESS at home, and likewise the Stax SR-X9000 with Stax SRM-T8000, on one occasion driven by a streamer and on another by my FiiO K9 Pro ESS's line out. They are both exceptionally clear, but I had already refined upon that with EQ for the Meze Elite which probably also had a cleaner CSD, and they both in the absence of crossfeed still imaged either directly from the drivers or headphone cups or along a small arc in front of my face. I am already able to experience exquisite texture and incisive transients through my FiiO K9 Pro ESS driving the EQed Meze Elite or HE1000se with the right recordings, but I can still hear points of discontent relative to the vivid goal.
• Some folks call it "holographic" when sounds image from above or behind the head or from "interesting and unexpected locations", but to me, that just sounds like imaging errors relative to a proper stereo mix unless the mixer did in fact employ binaural manipulations. Only DSP has projected the sound accurately and coherently (i.e. different frequency bands for the same instrument should not be imaging from different locations) on a stereo line a meter before me. If you believe a certain DAC or amp combination can somehow cause my headphones to image the music ahead of me similar to speakers without crossfeed, go ahead and recommend it.
• I own some "fancy" cables (see my signature) purely for the look and feel on top of their providing 4.4 mm balanced terminations, and for the opportunity to measure them, my personally experiencing firsthand that the only benefit was not having to turn up my DAC/amp as much for the same volume. See https://www.head-fi.org/threads/my-...and-a-copper-one.970695/page-23#post-17891886 (post #345).
• Of course, not all recordings are made equal, though Idagio (subscribed with FLAC for those who care) has done well to expose me to plenty of excellent ones in regard to clarity, vividness, and vastness. I have found that woodwinds have been the most likely to sound quite close to the live sound except in "stage depth", whereby it has mainly been strings that I just want to sound more real (though there have been times when I felt like I preferred the recorded sound of strings to the live sound). Perhaps I am hitting a limit related to the recording technology and techniques where no EQ or "playback transparency improvement" will fix a recording that did not capture the specific harmonic distribution that reached the audience, or that some spatial depth within the recording may not actually be captured or can indeed only be rendered illusorily. I have yet to try Qobuz which may or may not be as nice for finding and comparing many different recordings of the same piece.
• I am using a Windows system with Voicemeeter feeding audio via ReaRoute ASIO into the Reaper DAW which performs the binaural DSP. I remain skeptical about claims of the superiority of streamers, power conditioning beyond that done by the gear itself, or upgrading one's USB interface if not relying on some other connection.
• While vividly reproducing acoustic music is the priority, I also do appreciate "highly technical" modern works including stuff by Susumu Hirasawa as well as Japanese indie folk vocals.
• I had lately spent 30 minutes at a shop with a FiiO K19 and found that it certainly didn't do anything "magical" with the DCA Stealth, Expanse, and my own HE1000se, the Expanse sounding no bigger than what I remembered of driving it with my FiiO K9 Pro ESS albeit 11 months prior (I didn't have a chance to directly A/B since I was busy doing in-ear measurements for my evaluation), and at worst having the DCA Stealth give me my first actual taste of a "synthetic, artificial, analytical, bright, cold sound" relative to my actual HRTF measurement EQs and binaural DSP for simulating studio monitors in an anechoic room.

Given what I have experienced with headphone transducers, decently high-end cables, estats in audio show or shop conditions (not too noisy), and my having not been so impressed by the XI Audio amps with Diana MRs at an audio show (they sounded equally clear, detailed, big, and textured after a cursory volume match), my objectivist skepticism is still open to sating the curiosity of whether or not my FiiO K9 Pro ESS could possibly be holding back my audio experience. As such, I happen to be willing to dish out a decent bit of moolah on a nice DAC and amp (and "fancy" interconnects so folks don't "cry foul" with my using Monoprice cables) to maximize the chance of being "convinced" of what the merits of R2R or class A may be, else to bust a myth for myself while still having gear that is nice to look at. So far, the Holo Audio May KTE and Bliss KTE have been the most appealing to me from both the measurements and looks perspective, though it may be impossible for me to audition such within Toronto, Ontario, Canada; the main shop that I know to be good for demos mainly features ESS-based or Chord DAC/amps other than the Naim Uniti Atom HE, and even if I were to book a listening session, I would feel bad about being unlikely to purchase their products, and my curiosity of the Holo Audio stack would remain. I am thus interested in what other stacks within this price range others would recommend based on the listening experience and expectations I have described, else if it seems like the way I hear things really is unlikely to benefit from purported R2R or class A etc. improvements such that it would be best to stick to improving my HRTF measurement and DSP results.

 

[dbull12]

I don't have personal experience with this gear, but since all your headphones are planars based on everything I've read I think Ferrum Oor + Hypsos would be a good alternative to the Holo Bliss since its specialty is driving planars. Then for DAC there's the Ferrum Wandla but that's also an ESS implementation so if you're looking for a taste of R2R then Holo Cyan 2 would be a good place to start to get a good sense of the Holo "house sound" for way less money (and way more space savings) than the May

Edit: Not sure of the likelihood of this, but I'm holding out hope that Holo will release a HP amp the same size and similar price as the Cyan 2, as I think that would make a really nice desktop stack that could be "endgame" for many like me

 

[bagwell359]

Interesting. Planars dating back to my first Magnepan MG-1 I bought in the Summer of 1978 always seem to do better with more power. It's true of my headphone experiences too. That would lead to recommendations of the Schiit Ragnarok 2, Burson Soloist, etc. I favor full Class A or sliding bias Class A with full balanced topology to give the fullest definition/stage depth/harmonic structure over other types such as AB, D, etc.

As for cables - except for corner cases - I'm good with O2 free capably constructed cable of the lowest price (or DIY). In headphones that corner case for me are HFM cables which somehow manage to actually sound poor in double blind testing. For interconnects I go for XLR. For headphone cable - XLR over SE, unless the amp puts out more power in SE for some reason.

For DAC's given my exposure - which I just outlined in the Aune S17 Pro thread, I'm a strong proponent of R2R DAC's as having the ineffable dimensions that recorded music so often lacks over D-S DAC's.

I was weaned on the BSO, and my reference generally is live unamplified music which is similar to you. I appreciate that science/engineering is needed to create musical playback systems, and also that the experience itself is fully subjective - although like wine tasting one can educate ones ears.

I do appreciate the diligence of your efforts. Please let us know what you come up with!

 

[MrHaelscheir]

Interesting. Planars dating back to my first Magnepan MG-1 I bought in the Summer of 1978 always seem to do better with more power. It's true of my headphone experiences too. That would lead to recommendations of the Schiit Ragnarok 2, Burson Soloist, etc. I favor full Class A or sliding bias Class A with full balanced topology to give the fullest definition/stage depth/harmonic structure over other types such as AB, D, etc.

As for cables - except for corner cases - I'm good with O2 free capably constructed cable of the lowest price (or DIY). In headphones that corner case for me are HFM cables which somehow manage to actually sound poor in double blind testing. For interconnects I go for XLR. For headphone cable - XLR over SE, unless the amp puts out more power in SE for some reason.

For DAC's given my exposure - which I just outlined in the Aune S17 Pro thread, I'm a strong proponent of R2R DAC's as having the ineffable dimensions that recorded music so often lacks over D-S DAC's.

I was weaned on the BSO, and my reference generally is live unamplified music which is similar to you. I appreciate that science/engineering is needed to create musical playback systems, and also that the experience itself is fully subjective - although like wine tasting one can educate ones ears.

I do appreciate the diligence of your efforts. Please let us know what you come up with!

Do you have much experience with the Holo Audio products? Other than the looks and internal build, the appeal is claims of achieving the merits of R2R and Class A without introducing any concerning distortion (as corroborated by measurements). The May KTE is particularly attractive per the ability to do much more advanced digital filtering upstream (which if I understand correctly could compete with the reconstruction filters in Chord products or the Ferrum WANDLA); I am nonetheless curious of any other models that could appeal. My personal experience with distortion at least when introduced digitally (Waves Tube Saturator Vintage, PK Harmonic, or using SocaLabs Maths to implement custom transfer functions, my using a digital oscilloscope in the Reaper DAW to confirm that no digital clipping is happening and an FFT analyzer to look at the distortion profile for tones) is that it is either inaudible, or by the time it starts getting audible, it soon comes to make the dynamic peaks sound "horrific". You may also be familiar with https://www.klippel.de/listeningtest/, but I will accept other factors contributing to people's subjective experience, such as actual tonal changes with tube amps or anything with a higher output impedance. While folks may find the "SINAD wars" excessive, one must note that those numbers are only achieved at full scale, there maybe being concerns if driving the DAC or amp at lower levels where the SINAD decreases; I ideally want to see if a higher-end or highly different implementation achieving comparable distortion performance to my FiiO K9 Pro ESS with internal THX 788+ could possibly sound different.

I am also supposing that it would be a better idea to first drive a Class A amp with an ESS DAC than to skip to using an R2R DAC to drive a THX amp.

For cables, at least for headphones, my findings can be seen below including more recent HiFiMan cables, but that won't stop people from hearing what they hear, but I know what I personally don't hear. Worst case, one can claim my FiiO K9 Pro ESS's own "harshness" if any was "masking" that of the HiFiMan cables.

 

[bagwell359]

Do you have much experience with the Holo Audio products?

Not any, but the reputation/Technical supporting information seems to be of a high order.

ther than the looks and internal build, the appeal is claims of achieving the merits of R2R and Class A without introducing any concerning distortion (as corroborated by measurements). The May KTE is particularly attractive per the ability to do much more advanced digital filtering upstream (which if I understand correctly could compete with the reconstruction filters in Chord products or the Ferrum WANDLA); I am nonetheless curious of any other models that could appeal. My personal experience with distortion at least when introduced digitally (Waves Tube Saturator Vintage, PK Harmonic, or using SocaLabs Maths to implement custom transfer functions, my using a digital oscilloscope in the Reaper DAW to confirm that no digital clipping is happening and an FFT analyzer to look at the distortion profile for tones) is that it is either inaudible, or by the time it starts getting audible, it soon comes to make the dynamic peaks sound "horrific". You may also be familiar with https://www.klippel.de/listeningtest/, but I will accept other factors contributing to people's subjective experience, such as actual tonal changes with tube amps or anything with a higher output impedance. While folks may find the "SINAD wars" excessive, one must note that those numbers are only achieved at full scale, there maybe being concerns if driving the DAC or amp at lower levels where the SINAD decreases; I ideally want to see if a higher-end or highly different implementation achieving comparable distortion performance to my FiiO K9 Pro ESS with internal THX 788+ could possibly sound different.

You have delved and developed understanding in these areas that's impressive and well past me.

I am also supposing that it would be a better idea to first drive a Class A amp with an ESS DAC than to skip to using an R2R DAC to drive a THX amp.

Not sure about that. I have a sliding bias Class A amp which is particularly good with planars, and by today's standards an older R2R DAC.

For cables, at least for headphones, my findings can be seen below including more recent HiFiMan cables, but that won't stop people from hearing what they hear, but I know what I personally don't hear. Worst case, one can claim my FiiO K9 Pro ESS's own "harshness" if any was "masking" that of the HiFiMan cables.

My experience at a serious level is with the stock cable of the HE-500 (quite old), and the more recent HE-6 SE (surgical tube cable). Having abandoned the cable wars around 1995 as snake oil (minus a very few corner cases) I was shocked when I tried an alternate to the HE-500 cable. I couldn't believe it, I cut off the ends and rebuilt it with new connectors, I did extensive double blind testing and it was microphonic and splashy in the treble - disturbingly so. Later I has similar experiences with the HE-6 SE, but not quite as extreme. I've had no issues with DCA and Audeze cables vs aftermarket cables - double blind and not.

One issue which you may find interesting is CSD with planars. With older and/or cheaper planars easier to read on a graph/hear is the CSD behavior. Planars react to input signals similar to ribbons or estats and I assume AMT technology too, but, when the signal is off planars do in general stop moving later. Now at -40 db under the signal its not supposed to be audible, but with multiple signals and harmonics decaying at different times/rates in complex music on very excellent gear there is something remaining - a haze is one way I could characterize it. In my case with damping of the waves off the stators, dynamat to curtail ringing of the cup, open rears, different pads, PEQ slightly cutting areas of high CSD - the overall result is a cleaner sound - but when compared to the DCA estats or Rall ribbons (with filters) in direct comparison that sense of haze is again detected.

I do think planars is the most cost effective technology in headphones, and the ear is much more sensitive to the leading edge of the note than the trailing parts - but you may also want to try ribbons or estats some more to see if they may bring you closer to the goal. I know that for me with loudspeakers - ribbon tweeters are my overwhelming choice, and that the two DCA estats, and the two Ralls I know (CA-1a and SR-1b) are the very best headphones in capturing the best recordings - with the Stax 007 mk2 being very close to those - sadly except for the Rall CA-1a they are well out of my price range. Of course the problem with all of those is that the bass doesn't have the impact/dynamics of the best planars/dynamics. But the DCA's in particular are decent at that along with the timbre which is excellent. The Rall CA-1a does a bit of HD/poor damping in the bass which is the weak point, but over about 320 Hz - on top end equipment and with the filter it tops the Susvara/D8000/LCD-4/HE-1000 SE for me.

 

[Whazzzup]

You lost me at stack, except convenience or affordability

And yes a music server is probably the most important piece in an audiophile rig.

Good luck use each company for what part of the chain it does best

 

[MrHaelscheir]

Not any, but the reputation/Technical supporting information seems to be of a high order.

You have delved and developed understanding in these areas that's impressive and well past me.

Not sure about that. I have a sliding bias Class A amp which is particularly good with planars, and by today's standards an older R2R DAC.

My experience at a serious level is with the stock cable of the HE-500 (quite old), and the more recent HE-6 SE (surgical tube cable). Having abandoned the cable wars around 1995 as snake oil (minus a very few corner cases) I was shocked when I tried an alternate to the HE-500 cable. I couldn't believe it, I cut off the ends and rebuilt it with new connectors, I did extensive double blind testing and it was microphonic and splashy in the treble - disturbingly so. Later I has similar experiences with the HE-6 SE, but not quite as extreme. I've had no issues with DCA and Audeze cables vs aftermarket cables - double blind and not.

One issue which you may find interesting is CSD with planars" With older and/or cheaper planars easier to read on a graph/hear is the CSD behavior. Planars react to input signals similar to ribbons or estats and I assume AMT technology too, but, when the signal is off planars do in general stop moving later. Now at -40 db under the signal its not supposed to be audible, but with multiple signals and harmonics decaying at different times/rates in complex music on very excellent gear there is something remaining - a haze is one way I could characterize it. In my case with damping of the waves off the stators, dynamat to curtail ringing of the cup, open rears, different pads, PEQ slightly cutting areas of high CSD - the overall result is a cleaner sound - but when compared to the DCA estats or Rall ribbons (with filters) in direct comparison that sense of haze is again detected.

I do think planars is the most cost effective technology in headphones, and the ear is much more sensitive to the leading edge of the note than the trailing parts - but you may also want to try ribbons or estats some more to see if they may bring you closer to the goal. I know that for me with loudspeakers - ribbon tweeters are my overwhelming choice, and that the two DCA estats, and the two Ralls I know (CA-1a and SR-1b) are the very best headphones in capturing the best recordings - with the Stax 007 mk2 being very close to those - sadly except for the Rall CA-1a they are well out of my price range. Of course the problem with all of those is that the bass doesn't have the impact/dynamics of the best planars/dynamics. But the DCA's in particular are decent at that along with the timbre which is excellent. The Rall CA-1a does a bit of HD/poor damping in the bass which is the weak point, but over about 320 Hz - on top end equipment and with the filter it tops the Susvara/D8000/LCD-4/HE-1000 SE for me.

In https://www.head-fi.org/threads/hifiman-he1000-se.886228/post-17865003 (post #4,788), you can see that I have taken quite an interest in CSD measurements (I have a better waterfall perspective in video footage or headphone pad measurements that I have yet to edit) insofar as I thought I could hear such residuals at least for extreme transients, though I had found that in using my own head as the coupler, I didn't get nearly as bad "planar ridges" as the folks on SBAF found, and I would also rather not have to make major tonal changes to mitigate large CSD issues which for my own measurements aren't actually that sharp. When I had the chance to measure them, the HE-1 and X9000 had surprisingly higher midrange to treble CSD residuals than my Meze Elite or HiFiMans, such most likely having been well above the background noise floor, though I at the time didn't have my piece of acoustic foam available to help absorb the back wave, but my living room isn't anechoic anyways. I had lately measured the DCA Stealth, Expanse, and E3 which were indeed comparably clean in CSD to my ATH-M50xBT, but were possibly outdone in multi-tone distortion by my EQed Meze Elite; I can't/shouldn't share the details yet. Maybe cleaner CSDs could have audible benefits for "resolution", but my findings didn't justify buying a DCA to find out, hence my now focusing on DAC/amp shopping, and if my measurements are to be believed, my EQed ATH-M50xBT may already be enough to simulate both the DCA Stealth or Expanse's CSD and multitone distortion. Estats still haven't done much for me compared to EQ in what listening opportunities I had, and I am skeptical of the current RAAL offerings, especially if they are hard to EQ, mind I like to have deep bass extension which classical can in fact benefit from

Anyways, back on topic, have you ever experienced a DAC or amp upgrade that made the difference of transforming "on or between the headphone drivers" sound to a consistently and convincingly out-of-head experience? I lately watched Soundnews' review of the Ferrum WANDLA and heard his description of the Rockna Wavedream purportedly transforming headphone listening into binaural or nearfield, which seems outlandish considering that even the HE-1 for me images identically to any headphone I've heard without crossfeed or personalized HRTF rendering. And if you have had such out-of-head experiences, was there a minimum "budget" for achieving such?

 

[bagwell359]

In https://www.head-fi.org/threads/hifiman-he1000-se.886228/post-17865003 (post #4,788), you can see that I have taken quite an interest in CSD measurements (I have a better waterfall perspective in video footage or headphone pad measurements that I have yet to edit) insofar as I thought I could hear such residuals at least for extreme transients, though I had found that in using my own head as the coupler, I didn't get nearly as bad "planar ridges" as the folks on SBAF found, and I would also rather not have to make major tonal changes to mitigate large CSD issues which for my own measurements aren't actually that sharp. When I had the chance to measure them, the HE-1 and X9000 had surprisingly higher midrange to treble CSD residuals than my Meze Elite or HiFiMans, such most likely having been well above the background noise floor, though I at the time didn't have my piece of acoustic foam available to help absorb the back wave, but my living room isn't anechoic anyways. I had lately measured the DCA Stealth, Expanse, and E3 which were indeed comparably clean in CSD to my ATH-M50xBT, but were possibly outdone in multi-tone distortion by my EQed Meze Elite; I can't/shouldn't share the details yet. Maybe cleaner CSDs could have audible benefits for "resolution", but my findings didn't justify buying a DCA to find out, hence my now focusing on DAC/amp shopping, and if my measurements are to be believed, my EQed ATH-M50xBT may already be enough to simulate both the DCA Stealth or Expanse's CSD and multitone distortion. Estats still haven't done much for me compared to EQ in what listening opportunities I had, and I am skeptical of the current RAAL offerings, especially if they are hard to EQ, mind I like to have deep bass extension which classical can in fact benefit from

Highly interesting. Ralls do not do bass the way you want - as you suspect. As for reasonable priced bass with timbre: HE-6 SE (either version) and DCA E2. I'm interested in the E3 - but hoping they make an open version as so far closed backs are not my cuppa.

Anyways, back on topic, have you ever experienced a DAC or amp upgrade that made the difference of transforming "on or between the headphone drivers" sound to a consistently and convincingly out-of-head experience? I lately watched Soundnews' review of the Ferrum WANDLA and heard his description of the Rockna Wavedream purportedly transforming headphone listening into binaural or nearfield, which seems outlandish considering that even the HE-1 for me images identically to any headphone I've heard without crossfeed or personalized HRTF rendering. And if you have had such out-of-head experiences, was there a minimum "budget" for achieving such?

Not to the extent you are looking for.

But big watts do increase sense of bass impact dynamics, instrument focus, and blackness of backgrounds with planars.

For 300 ohm headphones such as the HD-600 and 800 - OTL tube amps give truer timbre/density, more accurate harmonics, and slightly bigger slightly less damped bass (a mix of increased HD and voltage).

My DAC (Schiit Gungnir MB A1) chosen after painstaking comparisons with my $15k phono had better bass impact, timbre (not a surprise), equally wide and deep soundstage, and timbre which certainly way outstripped any D-S I've heard and being quite similar to the Krell 20i of 20 yrs ago - the first player that ever sounded like music to me.

 

[MrHaelscheir]

I would like to share my recent experience A/Bing a few amps and DACs at two different shops.

Most listening was done with Qobuz which I had gotten a trial subscription for that morning and found to not at all be advantageous for the same recording compared to Idagio out of MS Edge including with high-res recordings. The Qobuz tracks were played at the highest resolution available. Sample tracks included:

• https://open.qobuz.com/album/0030911171520 (Thierry Fischer's Mahler Symphony No. 1; "Mahler 1"): Particularly the opening of the fourth movement for bigness and scale among other qualities, and around 13:10 in the first movement to the climax. This recording wasn't used as much as the others. All the symphonic examples also test for clarity, separation, and "vividness".
• https://open.qobuz.com/album/0002894534162 (Pierre Boulez' Mahler Symphony No. 5; "Mahler 5"): The opening tutties for bigness and scale among other qualities, and the opening of the second movement for texture and bass impact.
• https://open.qobuz.com/album/bgigk6cvwdfwb (Herbert Blomstedt's Brahms Symphony No. 3; "Brahms 3"): Particularly the opening of the first movement for texture and scale, 6:29 for texture and string tactility, and various parts of the fourth movement including around 1:54, also particularly focusing on string texture and I suppose indeed any differences in "epicness" as a more general term.
• https://open.qobuz.com/album/f11c5m67o6jna (""gift" at Sogetsu Hall (Live)" by Ichiko Aoba): Particularly the fourth track, "Kokoro no Sekai", for vocals and the details of guitar attacks.
• The rather nice recording below ("zhongruan concerto") for imaging of various instruments throughout the stage, and for zhongruan pluck details, weight, and incisiveness.
  
  

As for hearing acuity, I will say I am in my mid-20s with my last audiometry session in September having shown no issues, IIRC a threshold around or below 5 dB, and for a DIY check with a rough 85 dB (through my Latnex SM-130DB SPL meter, which I admit is pretty loud) at 1 kHz reference with my "V3.1 PEQ" Meze Elites (see later) perceive tones up to 20 kHz despite my tinnitus, or up to 19 kHz through a Genelec 8341A for a 75 dB at 1 kHz reference, though this is more susceptible to comb filtering.

First shop:

Figure 1: Listening room setup. I wasn't the one who stacked the DAC/amps. Personal FiiO K9 Pro ESS with my own Meze Elite Tungsten with premium silver-plated PCUHD upgrade cable with 4.4 mm termination. The Meze Elite as seen in https://www.audiosciencereview.com/...n-susvara-headphone-review.50705/post-1888972 (post #1,183) is the lowest bass and multi-tone distortion headphone I have ever measured, probably rather well competing with DCA's offerings (when EQed) and being my preference for comfort, looks, and presentation.

The source was my Lenovo Yoga 900 laptop with the charger disconnected since I had unfortunately forgotten such, mind didn't want any contention between the K9, K19, and cellphone USB tether connections. I have a separate desktop at home for primary listening. In all cases, the "V3.1 PEQ" (the seventh figure in https://www.head-fi.org/threads/mez...eadphone-official-thread.959445/post-17743502 (post #5,152), but with more 6 kHz and less 1 kHz so as to be closer to my measured HRTF) as processed via Equalizer APO was used; Equalizer APO should be transparent, introducing no distortions of its own except related to increasing the amplitude of a frequency band, its minimum-phase operation only working to correct phase errors toward linearity.

The FiiO K9 Pro ESS was on low gain with the volume knob around 11 o'clock for the DAC/amp's vertical orientation (2 o'clock when laid on its side) for the symphonic works.

All the available DAC/amp combos at this shop in this setup used ESS DACs, so except for the FiiO K19 which had no line inputs, I felt no point in trying to listen out for differences in ESS DAC implementations, my driving them with my FiiO K9 Pro ESS's dual ES9038PRO DACs' line outputs. Here, I used my premium silver-plated PCUHD upgrade cable with 4.4 mm termination for those amps that could accept such, and borrowed a stock single-ended 1/4" cable for those amps that did not have a 4.4 mm output. I do not hear differences between unbalanced and balanced connections when volume-matched.

I had not read or watched anything about the claimed "sound signatures" of these amps other than a recommendation of the Sennheiser HDV 820 and the claim of the FiiO K19 having more detail and a bigger soundstage than the FiiO K9 Pro ESS, which to me is an immediate sign of a failed volume match.

First up, XLR line outs of my DAC/amp sent into the XLR line ins of the Sennheiser HDV 820 via a pair of 3' Monoprice Premier XLR cables, whatever you may think of those interconnects which for all I know should be transparent. The volume knob's resistance is on the higher side of things, whereby any more resistance and I would be comparing it to the SPL Phonitor amps I encountered at an audio show, and I absolutely hated the feel of those knobs. On first impressions, nothing sounded special. Now, I acknowledge that this amp was designed with a higher output impedance and wasn't meant for lower (mind mostly flat per https://www.soundstagenetwork.com/i...e-audio-elite-headphones&catid=263&Itemid=203) impedance headphones mind measurably exhibits worse distortion for such, but the point is that even then, for my test tracks, when employing a simple volume match, the HDV 820 did not exhibit any differences in soundstaging, imaging, detail and texture for strings, bass impact, brass texture, plucked weight, and transient incisiveness. None of these amps or DACs made a difference in the flanking instruments of symphonic recordings imaging directly from the drivers. The volume knob was a bit past 12 o'clock for the symphonic works.

It was the same deal with the FiiO K19 which I had set to "High Gain" with a volume of 115 out of 120 for the symphonic works. Effectively identical everything from tonality to texture and all when volume matched. Now, I had forgotten to bring one of my HiFiMan cables to expose terminals for using my multi-meter, but in going back and forth with small volume matching adjustments, any perceived differences would eventually disappear, or sometimes I may have even thought I had a subtle preference for the FiiO K9 Pro ESS.

As a reference, this is how I hear the imaging in the "zhongruan concerto" track through the Meze Elite without EQ or DSP:

• By 14:37, a shaker starts up from the far left direction and a bit ahead, not imaging from the drivers themselves as with most far panned music for me, but in an area just beyond the furthest I can direct my gaze leftward. With my speaker simulation DSP, it images a bit to the right of the left virtual channel.
• The zhongruan (soloist's plucked instrument) images a bit toward the right. Sometimes it can feel like it is upwards of a meter ahead like with my Genelec 8341As or my binaural speaker simulation, but on closer analysis, it is rather imaging an inch or so in front of my forehead. With my speaker simulation DSP, channel balance corrections may have allowed the zhongruan to be more centered, but it most notable images straight ahead and coherently, not too high or low or anything.
• Background bass notes image ahead. I may have not paid as much attention to those during the auditions.
• At 14:46, background plucked instruments commence ahead around the middle left; I technically didn't make note of this during the audition. With my speaker simulation DSP simulating speakers 30 degrees left and right, these instruments imaged around 20 degrees to the left.
• At 14:56, more prominent plucked instruments image from ahead center and toward the right, at the extremes wrapping around toward my headphone's right driver. I hadn't paid much attention to the resuming background plucked strings on the front left. With my speaker simulation DSP, they also image in multiple distinct points, but clearly ahead along a stereo line, neither too high nor too low (mind I can adjust said height).
• At 15:04, plucked string tremolos are presented right ahead (I would later on realize they were actually or also imaging high up; again, my ears tend to cause upper treble through headphones to image almost above my head) with sharp attacks speckling the stage with some ambient air, being a nice case of this recording encoding some sense of stage depth. My ears without binaural rendering also tend to image treble to upper treble content a bit high rather than exactly straight ahead. With my speaker simulation DSP, they image from a line around the front left.
• At 15:08, flutes play from the front left, maybe around 35 to 40 degrees left whereas the shaker images maybe 80 degrees left, their also having some ambient air. With the speaker simulation DSP, they image around 18 degrees left.
• At 15:14, the flutes on the front left make another flourish, and by 15:15, reed instruments pulse from the front right around 30 to 35 degrees right. With my speaker simulation DSP, the flutes image 20 degrees to the left and the reed instruments right from the right virtual channel 30 degrees to the right.
• At 15:28, the erhu pizzicatos image from the front left with some pizzicatos from a section on the further front right. With my speaker simulation DSP, they image from areas 20 degrees left and 25 degrees right.
• At 15:32, these plucked strings image from right ahead, though on closer inspection having some higher frequency content imaging like 60 degrees high; this height issue is not present with my speaker simulation DSP. There is pizzicato from the front left and further front right, the right plucks coming closer toward the right driver; with my speaker simulation DSP, these again image 20 degrees left and 25 degrees right.
• At 15:36, the lower-register plucked instrument images directly from the right driver. With my speaker simulation DSP, this images right from the virtual right channel 30 degrees to the right.

This imaging presentation was identical for my ears through all these amps and likewise the DAC/amp combos heard at the second shop. There were absolutely no at least directional or distance imaging differences to be heard for my ears and listening methods.

I had then switched to the borrowed stock unbalanced 1/4" cable for the Meze Elite to use with the McIntosh MHA200 tube amp which I drove via the XLR inputs. This tube amp had already had around an hour to warm up. Switching the load impedance knob on the left from 32 Ohms to 300 Ohms or 600 Ohms simply increased the volume, whereby I suppose this was simply increasing the voltage gain. The volume knob on this demo unit was a bit wobbly I suppose after lots of use, and likewise had a lighter but for the most part smooth resistance, rather light for my preference considering the steep changes in volume with small motions, having a 12 o'clock detent which I would rather not have; my volume setting was around 10 to 20 degrees to the right of 12 o'clock. Otherwise, I heard absolutely no tonal differences (McIntosh themselves advertise flat response with no more than -0.5 dB deviation in the audible band) or "tube sound", any tube distortion having probably. been inaudible at this operating point. At one point shortly thought I heard "tube sound" from Ichiko Aoba's vocals, but that was just a property of the recording itself, any warmth, fullness, and vocal texture being presented largely the same between these amps. Details, texture, imaging, "bigness" with symphonic works, bass impact or dynamics, string pluck weight or incisiveness and all were effectively the same compared to my FiiO K9 Pro ESS.

Figure 2: The Moon 230HAD was hooked up to my FiiO K9 Pro ESS's line out using the shop's AudioQuest Golden Gate RCA interconnects into the A1 input.

The volume knob maybe had slightly less resistance than the Sennheiser HDV 820's. Otherwise, again, this amp sounded essentially identical to the others when volume matched no matter how closely I listened. If ever I thought I was starting to notice a difference in clarity or presentation or something subtle with tonality or the holistic sound, it was most likely too subtle to matter, or that difference would eventually disappear.

I did not bother or have the time to listen through the EverSolo DMP-A8.

Second shop:

Figure 3: Setup with my FiiO K9 Pro ESS, Chord Hugo 2, and RME ADI-2/4 Pro SE.

Figure 4: My Meze Elite had looked nice in the ambient sunny daylight.

Here, I had meant to compare the DAC/amp combos as a whole. At this point, I was dismayed to find that my laptop's battery had died during the drive. The shopkeeper was against using EQ when evaluating gear under the unsubstantiated belief that software EQ is necessarily inferior to hardware EQ, but with my being forced to run Qobuz off of my cellphone, I would now have to do the evaluation with the stock Meze Elite hybrid pad tuning and no DSP. One thing I can say is that this stock tuning was at the time surprisingly agreeable insofar as I hadn't listened to any other tonality in that session, string texture coming through just fine, the mids indeed being forward, but in a full manner that was neither bloated nor dull, indeed somewhat "romanticizing" things, though also giving me the feeling of what it must sound like to be "inside a vacuum tube".

When I initially played some Brahms through the Chord Hugo 2 (which was my main target for this particular visit, whereby had it impressed me, I would have happily fetched the Chord Hugo TT 2 from this dealer), it simply sounded familiar and not particularly special, which was confirmed once I got my USB connection to the FiiO K9 Pro ESS working after realizing it was still set to output to the line outs. The Chord Hugo 2 sounded identically detailed and textured to the FiiO K9 Pro ESS for me, and also for the reference of having heard live orchestral performances many times with the partial aim of critical listening (the resolution of live sound is not infinite, unless you were to blame my mild tinnitus or my ears somehow distorting more than golden ears'; I can tell when a piece of gear is not bringing me any closer to the live sound I am looking to reproduce). Again, if ever I thought I heard the slightest overall difference, it didn't last long or could have probably been attributed to an imperfect volume match. This unit at the time was unfortunately low on battery. Nonetheless, it also did nothing special for me in the imaging of that zhongruan concerto track.

The shopkeeper was of the expectation that the RME ADI-2/4 Pro SE (dual ES9038Q2M) should with its more powerful amp (3.4 or 3.6 W into 32 Ohms compared to the FiiO K9 Pro ESS's 2.1 W) definitely sound more detailed and dynamic with a larger soundstage than the FiiO K9 Pro ESS. This was not the case for my ears and how I listen to and compare tracks, my first getting the gear to output at very similar volumes, my making note of the qualities of the sound through chain A, playing the section multiple times to try to best internalize that general quality, my then moving over to chain B and seeing how it compares; most of the time, it is simply like restarting the exact same sound, and in such case, I repeat the process of repeatedly listening to the same track to internalize the qualities before switching back to chain A, maybe sometimes for honing in doing more frequent switches. The shopkeeper was of the belief that you should normally first listen to one piece of gear at a time to "get used to or learn its 'sound signature'" before then "comparing" it to the original piece of gear you had been listening to for the longest and "know" the best. My argument is that this approach is just asking you to psychologically prime a preconception of a sound, possibly even mentally encode a "mental EQ/DSP" that gets automatically applied when you switch back to the given amp or DAC in sighted listening. Sure, I could probably by random chance or letting my mind go loose hear a "difference" and let my mind reinforce such through confirmation bias rather than questioning that difference and very carefully checking to see if it is possible to perceive the same qualities through the other piece of gear, but that style of listening would to me be the complete opposite of "critical listening". In the ideal case, I would have both chains playing simultaneously with the capacity for instantaneous volume-matched switching, whereby if I don't hear the soundstage instantly expanding like I could simulate with A/Bing my speaker simulation DSP or a veil of detail being instantly lifted, any subtle differences if any just aren't worth it.

Anyways, there was a point where with the aggressive part of first movement of the Mahler 5 recording after 38 seconds, it was starting to seem as though the RME was exhibiting a slightly different character, a possibility of improved clarity, maybe indeed strong dynamics, and particularly a tighter and slightly more detailed bass in that aggressive throb where it was somehow seeming to consistently sound duller through the FiiO K9 Pro ESS, as though the RME was bringing out a bass detail I had not heard before while the FiiO seemed duller or less textured there, almost as though I were for the first time hearing what “flabby bass” sounds like now that I had a “better” reference. I was starting to get excited since such a revelation of my amp actually being dynamically deficient there and losing this microdynamic detail would have single‑handedly made it worth pursuing the Holo Audio Bliss KTE for that "certainty" of performance for everything I throw at it.

Figure 5: Holding multimeter probes against the terminals of the Arya Organic's unbalanced cable.

I finally remembered that I had borrowed the Arya Organic with an unbalanced cable with which I could now have exposed terminals for doing a multimeter‑based volume match; I had opted not to listen to the Arya Organic lest I spoil my familiarity with the stock Meze Elite’s tonality with these tracks. For the Mahler 5 volume match, I registered 0.325 V out of the RME for the “‑20.5” loudness setting on my multimeter’s (old but decent SOAR Model 8050 with 9 V battery) 2 VAC scale, and 0.334 V out of the FiiO. It initially seemed like the difference was still there, then that sentiment started to disappear. I then A/Bed the zhongruan concerto, registering 0.100 V through the RME for “‑30.5” and 0.99 V through the FiiO, though I came to go back and only be able to achieve 0.105 V through the FiiO after finding that the volume had come to snap down to the previous increment, only before said correction having had the RME sounding a bit more incisive for zhongruan pluck transients amid its around 2 dB advantage; the imaging was otherwise identical. Then in returning to Mahler 5, the voltages were instead 0.337 V and 0.330 V for the RME and FiiO respectively, whereby on starting with the FiiO, it was now the FiiO that had that "extra bass detail", whereby on switching to the RME, it was now it that sounded a bit duller or smoother just like the FiiO had previously, then on returning to the FiiO, it now sounded identically smooth. If anything, my mind had simply become primed to notice a certain detail out of the same presented sound out of one piece of gear and suppress that perception through another. There were no longer any reliable differences between these DAC/amps.

Conclusion:

For my ears and listening methods, I did not hear any meaningful or reliable differences between these amps and DACs, their being tonally identical, imaging identically, and presenting the exact same amount of abundant information out of recordings. This then begs the question of whether or not it would be worth trying to see if a higher-end Class A amp or R2R DAC could possibly make a difference for my listening methods and expectations. I did at least this last Sunday attend a wonderful performance of Dvořák's Cello Concerto where I heard wonderful textures, richness, "open brightness", and a specific tonality or harmonic distribution that I still yearn to be able to hear through a recording through headphones; my Genelec 8341As at least sans room treatment don't achieve it, my stock Meze Elite or HE1000se sans DSP don't achieve it, my speaker simulation DSP EQed best to simulate a perfectly neutral HRTF still sounds a bit tonally veiled by comparison, or the imaging is limited by the conditions of the HRTF measurement and the binaural rendering software. Maybe a revelation does still lie ahead with upgrading from the FiiO K9 Pro ESS, else it lies in acoustic panels plus their facilitating improved DIY HRTF measurements and my trying to code up a better HRTF acquisition and rendering solution.

 

[Gennadi]

По моему мнению, идеальный комплект (цена-качество) таков:
1. Вход от компьютера > хороший и недорогой USB-кабель от Supra 60-70$ > ddc Gustard U18, Singxer Su-6, Topping U90, MATRIX 3 на Ваш выбор (500-600$) > кабель HDMI Supra 100$
2. ЦАП Музыкант Пегас- 2 - 3500$
3. Усилитель Flux Lab Acoustics MENTOR 1230$ + КАБЕЛЬ ПИТАНИЯ 100$
4. Кабель межблочный XLR TOPPING
5. кабель для наушников FIIO LL- RD 2024 1,5м 210$