Magni 3 Heresy and 3+ Blind Listening

Discussion in 'Blind Testing and Psychoacoustics' started by purr1n, Dec 8, 2019.

  1. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,772
    Trophy Points:
    113
    Location:
    Padre Island CC TX
  2. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,772
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Next steps:
    • Hide all the crap so I can't which wire goes where.
    • Turn on and wait a few minutes to let 3+ operating points reach equilibrium.
    • Have rest of family swap or not swap phono jacks among the amps when I'm not in the room.
    • Level matching madness down to the 1/1000 Vrms (the fancy test board will then be removed - the phono jack of the headphone will be plugged into directly the output breakout cables from the switcher)
    For amp #1
    IMG_20191208_131148.jpg

    For amp #2
    IMG_20191208_131157.jpg
     
  3. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,652
    Trophy Points:
    113
    Location:
    On planet
    This is my ABX system:
    20191205 Magni 3+ Heresy ABX.jpg

    It appears in the Magni 3+ vs Heresy thread but I would like to include it here as well.

    There are no time limits on the test runs, which is very important. Quick change overs do not work well, rather it takes time on each of A, B and X to be successful. The system makes switch over convenient so the listener may stay focused rather than be distracted by housekeeping chores and cable changes.

    Level matched to 0.05 dBu with Keysight 34465A and schiit-load set to 300R for HD800-Jmod listening.
     
  4. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,772
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Additional Notes:
    • Will use ZMF Verite where impedance is 300-ohms.
    • 0.455V is about 0.7 mW at 1kHz for these headphones. That approximately is 95db SPL give or take a few db. ZMF specs indicate 97db/mW. My measurements say it is a bit more efficient than that, but generally confirms that number isn't way off.
    • The level-matching was performed with a full scale signal from the DAC. For singular test tones, this level is fricking deafening! For music, it's moderately loud for me. In any event, I wanted to establish 0.455V or 0.7mW as the full scale peak.
     
  5. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,772
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    I will be not be doing rapid A / B switching. Instead I will listen to one, pause, and then listen to another. There will no listening time limit and I will be allowed to switch at any time. In addition, I will use using a DAC that I am familiar with, the Modi 3. Recordings will also be ones I am familiar with.

    Both amps will be set to high-gain. x5.5
     
    Last edited: Dec 8, 2019
  6. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,772
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Ugh. In terms of being able to tell the difference, this is too easy.

    The headstage gives it away. I can actually do rapid-switching here. One amp stages much closer with the headstage pushed to the plane of the ears (more inside the head) and pushes instruments near the 1/3 to 1/2 near the edges even wider out with a more of a hole-in-the-middle effect. The other amp has a headstage slightly further out. The stage is just as wide, but the instruments sound more evenly spaced. There feels like these is a better center to the stage. The differences are small, but they are there.

    Tip: Listen to the L-C and R-C panning of "c'mon c'mon c'mon" of Lose Yourself To Dance. Switch while in the middle of the panning effect.
     
    Last edited: Dec 8, 2019
  7. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,652
    Trophy Points:
    113
    Location:
    On planet
    My statement of methodology was generalized and intended for the greater audience, especially non-members. Many of my posts are written as if they were displayed on billboards all around the world, which they are.

    I know you are more than competent and careful at this.
     
  8. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,772
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Checked out Masiky and Rostropovich Bach Cello Suites:

    One seems to be simpler and cleaner, less busy. In the sense of missing softer level sounds in the midst of louder sounds, in the overtones of the Cello, the nature of the noise on the tape. The center image wavers more depending upon the string being played. The other one has greater richness and complexity to the sound, filling in gaps that the ear and brain are expecting. The micro-waverings in volume and frequency of the strings are better captured. This last one would be understood by those who have played string instruments.

    Tone and transient response are extremely similar between the two. Unsure if I can pick these out if these two aspects were the only ones.
     
  9. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,772
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    HYUNA Roll Deep:

    Not sure. Maybe, maybe one (the one which usually has the less wider and slightly deeper headstage) has just a bit more sibilance. With this track, can't even tell in terms of headstage, imaging is weird as if no effort was made in the studio to at least fake something. These k-pop recordings are loud and crushed and sounds are so processed, so can't rely on microdynamics or timbre. Wouldn't bet even $10 with this track. I could be fatigued.
     
  10. allegro

    allegro Friend

    Pyrate
    Joined:
    Nov 4, 2017
    Likes Received:
    637
    Trophy Points:
    93
    Location:
    Florida
    If staging is different among the amps would not it make a blind test more easy and compelling on speakers rather than on phones?
     
  11. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,772
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Yes. Maybe.

    I'm not a headstage person, but I've learned how to be one. Human beings are really good at perceiving tiny differences once they zoom-in and with practice.

    For me speakers is easier. It's how I listened when I grew up, only using headphones with Sony Walkmans, the ones that took cassette tape. I've noted that folks who began their hifi journey with headphones are much more attuned to headstage.

    I joke that headstage in pointless because what we are taking about is if the stage at my ears, in my head, behind my eyeballs, or 1mm in front of my eyeballs. It's like comparing bicycles to cars, with the cars being speaker soundstage.

    But I get it. Many headphone listeners don't like the in the head effect, and if we can move the headstage 1/2 forward, the difference is huge. Heck, some prefer a super wide effect, which I happen to detest.
     
  12. bboris77

    bboris77 Friend

    Pyrate
    Joined:
    Dec 12, 2015
    Likes Received:
    778
    Trophy Points:
    93
    What about testing the amps in low gain? It would be very interesting to see whether the Heresy suffers less than the Magni 3+ in low gain mode considering it uses feedforward correction instead of current feedback. In my experience all Schiit amps sound better in high gain because of less feedback.
     
  13. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,772
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Will try in low gain and report back.

    I just checked which button belonged to which amp. It's as I expected. I'm sure most of you can guess which one was which. Will ask kids to keep scrambling the phono plugs in the next few days.

    At high gain, the Heresy did much better than expected. These amplifier chips have come a long way since the BB BUF634 or that ubiquitious TI part (found on that Sennheiser HDVD800 headamp).
     
    Last edited: Dec 8, 2019
  14. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,772
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    As far as headstage, let's take a look at crosstalk. I will borrow @atomicbob's measurements for this:

    Magni 3+
    upload_2019-12-8_18-5-17.png

    Magni Heresy
    upload_2019-12-8_18-6-46.png

    We'd expect the Magni 3+ to maybe do worse in crosstalk because of the less wide and more centered imaging. However the opposite appears to be the case. Crosstalk performance appears to be fairly even within a few db from 30Hz to 2kHz. After 2kHz, crosstalk gradually increases on the Heresy to -75db at 20kHz.

    Looks like this measurement probably isn't showing us anything that may reflect what we perceive. It's unlikely that humans will be able to hear crosstalk -75db down at 20kHz and -80db down at 10kHz. That would be like hearing the slight hiss from desktop self powered monitors while being at a heavy metal concert where the sound guys crank up the volume to ear damaging levels.
     
    Last edited: Dec 8, 2019
  15. jexby

    jexby Posole Prince

    Staff Member Pyrate Contributor
    Joined:
    Sep 28, 2015
    Likes Received:
    8,145
    Trophy Points:
    113
    Location:
    Exit stage left....
    ^-- this defines me.

    while no blind listening with 3+ and Heresy, another delta during side by side today which tipped in favor of 3+ was the sound of cymbals. with Heresy, cymbals just sounded thinner with less metal and more sheen.
    and yes, less width in soundstage with Heresy as well.

    that being said, Heresy doesn't sound like a complete POS. which is good because the schiit name is on, and they are selling for (some?) meager profit apparently.
    IMO folks with less resolving cans may not hear/care about the subtle differences between 3+ and Heresy.
    deltas were much easier to detect with Auteur and less with Elegia.
    (this new Elegia is still burning in- but they are no plankton kings anyhow.)
     
  16. jexby

    jexby Posole Prince

    Staff Member Pyrate Contributor
    Joined:
    Sep 28, 2015
    Likes Received:
    8,145
    Trophy Points:
    113
    Location:
    Exit stage left....
    btw, the red LED on the Heresy drives me Fcking apeshit bonkers.
    looks like a complete fault / fail light where attention is needed to replace a broken hard drive or server failure.
     
  17. schiit

    schiit SchiitHead

    Pyrate
    Joined:
    Sep 28, 2015
    Likes Received:
    9,937
    Trophy Points:
    93
    Location:
    Texas and California
    Home Page:
    Without going all "I want to believe," there's a lot of anecdotal, subjective evidence that less feedback is better.

    You can take this to its logical extreme of "no feedback," which usually translates to "lots of local degeneration (feedback) to linearize the stages and get decent performance, but no overall feedback." We've done some amps like that. Pass and Ayre like to do amps like that. And, if you can get past the drawbacks of "no overall" feedback (high distortion, high output impedance, greater variability), that can be a great-sounding approach.

    But, then again, many times you can't get past the drawbacks, or (Heresy!) it turns out that adding feedback sounds better, or at least improves the subjective performance in ways that offset the downsides. So, it's not as simple as "less feedback = always better," no matter who is designing the amp.

    One thing's for sure: no matter the gain switch position, Magni 3+ has a whole heckuva lot less feedback than Heresy, which has about 100-115dB in the front end and 130dB in the output stage (read the datasheets for the OPA1662 and OPA1688, see open loop gain).

    speaker nosubtitle.jpg

    I'll leave you with this. Consider the above image.

    From one side, it could be read as "I want to believe...that there's more than just measurements to audio, because there are a lot of anecdotes beyond the analyzer."

    From another side, it could be read as, "I want to believe...that a limited number of steady-state measurements are the totality of what we need to understand the entire universe of audio."
     
    Last edited: Dec 8, 2019
  18. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,772
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    In low gain, the different in microdynamics and microdetail seem greater than in high gain. One amp simply sounds duller than the other and is missing ambient decays was well. On high gain, I didn't the microdetail difference between the amps were different with ambient decays, only with low level information while loud sounds were being played. The microdynamics difference on low-gain between the amps seems much greater.

    The electronic bass Yazoo's Only You and Situation have greater attack, greater heft on one amp. The other just kinds of plays it back in a robot like manner. It's kind of like at the dance studios where I see Asian kids do things perfectly, but somehow there is no soul in their movements because they are doing it because their Tiger-moms made them do it, and do it perfectly, not because they loved it. (Hahaha, I love being Asian because I can make these kinds of jokes).

    The decays and ambient sounds from the weird floaty keyboard beep and boop effects at the beginning of Situation are not as evident on the robotic sounding amp.

    The robotic sounding amp, which I am assuming is Heresy (my kids switched or maybe didn't switch the plugs), has less width this time around on low gain. The imaging seems more normal and less exaggerated toward the edges. This is different from high gain where the width was the same between both amps and the imaging pushed toward the sides.

    With the two tracks mentioned above, like high gain, I'm not hearing a difference in tonality that I would bet my money on, other than more low end heft from what I guessing the 3+

    That would be funny if I guessed wrong on low gain.

    --

    I don't think feed-forward has any effect on the effects of running things in a more open loop or less open loop manner. Again, feed-forward has been around forever and is nothing special. I've heard great amps using feed-forward (Hegel H2) and soulless amps (Benchmark AHB2) that I want to send @LetMeBeFrank to shoot up.

    I believe Heresy also uses feedback. Feed forward and feedback are not mutually exclusive.
     
    Last edited: Dec 8, 2019
  19. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,772
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    I guessed right. On low gain, the differences between the amps seem greater than on high-gain. I could also be more refreshed and less fatigued, as I had dinner, went out for a run up a hill, and hit the hot tub before this session.
     
  20. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,772
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Oh, proof that I level matched. Did I mention it's a major pain getting them exact to to the mV with those small volume control knobs?

    IMG_20191208_181843.jpg
    IMG_20191208_181833.jpg
     

Share This Page