Topping L30 Review and Measurements

Discussion in 'Headphone Amplifiers and Combo (DAC/Amp) Units' started by Marvey, Jun 28, 2021.

  1. atomicbob

    atomicbob dScope Yoda

    Friend BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    16,397
    Trophy Points:
    113
    Location:
    On planet
    I appreciate the intent of being fair, however 2V into an HD650 is 108.5 dBSPL and 4V is 114.4 dBSPL. Those are VERY LOUD levels. Transducer distortion will rise significantly, especially at low frequencies, certainly masking any amplifier distortion. ASR is plain FOS on this account. Denigrating amplifiers for distortion when clearly it is transducer distortion dominating audibility is a big detraction from credibility.
     
  2. mocenigo

    mocenigo Acquaintance

    Joined:
    Sep 26, 2017
    Likes Received:
    26
    Trophy Points:
    18
    Location:
    Munich
    Ok, I thought it was worse. I am not *that* tetchy.

    It's just that I did not understand this reaction to my post that was not meant to be polemic. Let me try to explain the reasoning behind my (admittedly, too succinct - to the point of resembling bait) comments.

    If an amp's frequency response measures flat, also way beyond the audible range, and the output impedance is also stable with frequency, then it cannot be dark or bright – this would imply that there is a change in frequency response. If the L30 paired to the headphones the OP was using sounds darker than other headamps this means either that 1) the other headamps were actually brighter in the sense that they emphasise somewhat the treble or that 2) the pairing makes the L30+those cans sound darker, perhaps by attenuating the treble, than the other headamps+the cans. One cannot draw a general and absolute conclusion on an headamp from circumstantial facts – and I am definitely not criticising the evaluation methodology of the OP which is absolutely correct, employing blind testing, and I believe he can hear what he claims to hear.

    Also, a lot of amps have ever so slightly increasing gain with the signal's amplitude (this is the effect of increasing distortion), resulting in higher perceived dynamics (macrodynamic, usually, but also microdynamics, resulting in mode perceived plankton, for instance). If the dreaded SINAD is very very high, this is a sign that an amplifier will sound dull in comparison. In fact, this may be one reason many people prefer slightly distorting amplifiers, as in Archimago's famous tests.

    I am actually in the investigation phase of the development of a piece of DSP software that will allow to create these effects in the digital domain. This, together with EQ, will allow one to get the desired sonic signature, dynamics, without having to purchase very expensive equipment.
     
  3. Thad E Ginathom

    Thad E Ginathom Friend

    Friend
    Joined:
    Sep 27, 2015
    Likes Received:
    11,276
    Trophy Points:
    113
    Location:
    India
    Good luck with that... because the audiophool masses will still prefer to buy that expensive equipment, especially arbitrary and quite-possibly-placebo stuff like cables. Fact is that a huge percentage of the audio market doesn't give a toss about measurements, because they think their ego can do better. The upside of that, is that they are not going to hanging on the words of Amir and ASR. The downside is, of course, audiophoolery.

    Anyway, actually good luck with the project. I always believed (now too deaf to hear it) that flat and transparent is good, but also that people are entitled to their personal taste is sound, whatever colour that is.
     
  4. Marvey

    Marvey Super Friend

    Staff Member Friend BWC
    Joined:
    Jun 6, 2021
    Likes Received:
    1,692
    Trophy Points:
    93
    No biggie. I'm glad we can clear this up. It's difficult to hear tone over the Internet.

    For sure. Although I wouldn't put it "slightly distorting" as a 1kHz tone is hardly a suitable tone and -85db THD+N is really fantastic when we really think about it. I think those who have worked in recording really appreciate -85db THD+N. Also, the mixing and mastering engineers aren't using 110db SINAD gear during playback for the artists or record company sign-off.
     
  5. Marvey

    Marvey Super Friend

    Staff Member Friend BWC
    Joined:
    Jun 6, 2021
    Likes Received:
    1,692
    Trophy Points:
    93
    Oh, I meant giving Topping benefit of the doubt. Although their close association with ASR makes this difficult. In this case of the borked right channel, I bet it's just a bad opamp.

    Don't get me started on ASR. The signature "addicted to fun and learning" pisses me off because there is no learning there. There's just dogma.

    BTW, this is for you @dsavitsk (your interest in back EMF) and everyone else here with respect to learning and discovery. Here is the same dashboard from the first post, but using a real HD650 headphone as the load instead of a resistor dummy load. Now this is very very interesting.

    Using HD650 as the load instead of a 300-ohm resistor
    HD650 in free-air, not on head
    upload_2021-7-1_7-58-19.png

    I did this twice to confirm. I still think I may be doing something wrong (checked switches to make sure I was not putting a parallel resistor load to the headphone. I gotta run, but will explore tonight. Don't come to any conclusions yet. Most times when stuff like this happens, I did something wrong. But if I didn't there could be something to back EMF.
     
    Last edited: Jul 1, 2021
  6. shotgunshane

    shotgunshane Floridian Falcon

    Staff Member Friend IEMW
    Joined:
    Sep 26, 2015
    Likes Received:
    5,158
    Trophy Points:
    113
    Location:
    Clear, clear water
    Back EMF with dynamic drivers is something @james444 and I briefly talked about a year or so ago, in regards as to possibly why low self powered and non-self powered dongles have limp or hazy bass with dynamic driver iems compared to a desktop amp with dedicated power supply. Something about higher current is needed to overcome the effects of the dynamic driver magnet. If this is a real and audible phenomenon, I’d like to understand more about it.
     
  7. Biodegraded

    Biodegraded Friend

    Friend
    Joined:
    May 28, 2017
    Likes Received:
    6,045
    Trophy Points:
    113
    Location:
    Vancouver BC
    Interesting! Was that with the headphones on a 'head', or in free air? Request: presuming you confirm this result, try both and see if there's a difference, and also try a higher synthetic load(s) corresponding to the approx HP load at 37 Hz to see what the distortion differences are. If the differences remain large, then back EMF (or some other electro-mechanical factor) must be at play.

    Edit: in red.
     
    Last edited: Jul 1, 2021
  8. Marvey

    Marvey Super Friend

    Staff Member Friend BWC
    Joined:
    Jun 6, 2021
    Likes Received:
    1,692
    Trophy Points:
    93
    Free air. I was dreaming about this last night so this morning I woke up and rushed to get some results. But yes, I thinking the exact same thing - do the measurements in free air and on head (coupler) because the impedance curves (Fs and amplitude of the bump) will be different.

    Well, I did mention the L30 sounded a "smeared" and lacked crisp transients in my subjective impressions. We'll see. I could have screwed up these measurements. Nine times out of ten, when I chase, it usually amounts to nothing.
     
  9. crenca

    crenca Friend

    Friend
    Joined:
    May 26, 2017
    Likes Received:
    3,400
    Trophy Points:
    113
    Location:
    Southern New Mexico
    This assumes the measurement captures everything. You cite yourself a situation (Archi's test) that confounds the logic. We all suspect there is more to it...
     
  10. mocenigo

    mocenigo Acquaintance

    Joined:
    Sep 26, 2017
    Likes Received:
    26
    Trophy Points:
    18
    Location:
    Munich
    Thank you for the wishes, and I agree 100% with you on personal taste.
     
  11. mocenigo

    mocenigo Acquaintance

    Joined:
    Sep 26, 2017
    Likes Received:
    26
    Trophy Points:
    18
    Location:
    Munich
    Well, I firmly believe that everything that can be heard (as in heard-heard, not heard-imagined) can be measured - our ears are not that precise an instrument. The question is whether we perform the right measurements. My personal opinion is that the only measurements we may lack is those that would capture hysteresis induced distortion. In other words, if you have not the dreaded SINAD but noise and distortion profiles, separated and represented as a function of frequency and of power, multi-tone tests (not only 2, but for instance the 32-tone tests), frequency response, phase, impulse response, and for instance a spectral representation of the combination of a low frequency loud signal superimposed to a high frequency weaker signal (to measure hysteresis induced distortion) then you are done.

    How they correlate to auditory perception is another matter, but a device is better if it performs better on the above measurements. Of course there are other factors, like the ability to drive (and not being driven) by downstream sinks. This can affect the behaviour in a different way than what the measurements would suggest, therefore once the measurements are done one must also determine the conditions in which they continue to hold (i.e. any difference w.r.t. the measurement is below the audibility threshold). This is going to be thorny.
     
  12. crenca

    crenca Friend

    Friend
    Joined:
    May 26, 2017
    Likes Received:
    3,400
    Trophy Points:
    113
    Location:
    Southern New Mexico
    I am tempted to say something sarcastic like "well you have it all figured out!" but you admit in the end it is a complex and thorny subject. I have objectivist tendencies as well - I mean a pressure wave (i.e. sound) is one of the simpler and well understood phenomena of nature is it not? Why should high (or even perfect) fidelity be all that complicated? If we can build an 787 surely we can.... Yet what exactly it is, let alone how to implement it in circuits and transducers remains elusive.
     
  13. mocenigo

    mocenigo Acquaintance

    Joined:
    Sep 26, 2017
    Likes Received:
    26
    Trophy Points:
    18
    Location:
    Munich
    I also started that part with "My personal opinion is ...".

    Well, recently Boeing has had their share of problems with their planes.

    I would say, however, that the transducers are the most difficult part of all.
     
  14. Psalmanazar

    Psalmanazar Most improved member; A+

    Slaytanic Cliff Clavin
    Joined:
    Sep 27, 2015
    Likes Received:
    5,037
    Trophy Points:
    113
    The distortion of the amp affects the lifespan of the drivers. The digital limiter in cheap crap is mainly there to protect the bargain basement chip amp from clipping and destroying the drivers with square waves. It’s a planned obsolescence accounting move to include it for less service calls. These are in the small JBL (Harman), Genelec, and Neumann (Sennheiser) he shills. These underpowered, high thd, amps are placed in the main cabinet of the speaker. They’re basically in an oven causing them to distort even more and cook the drivers. You can get better monitors that will cook themselves eventually for the same cost as their higher end models.

    Good new amps have never been cheap. The insane amps driving horns, Ns10s, and Proacs are not memes. The more reliable active speakers do not use the chip amps. They use massively overpowered tank amps shoved into a separate compartment onto the back of the speakers or bolted on with passive cooling. ATC and Quested use insane class AB amps, ME Geithains are the size of refrigerators, and Barefoot, HEDD, and Dynaudio now straight up shove high powered Hypex/Icepower/Pascal modules into their speakers without room correction. The room correction shit ASR faps over all runs out of headroom and will die like the Gainclone and STM Microelectronics bullshit.

    You know who doesn't fuck around at all? YAMAHA. Their shit lasts unless you blow the drivers. It's cheap but designed 100% the opposite of Chineseium.
     
  15. Marvey

    Marvey Super Friend

    Staff Member Friend BWC
    Joined:
    Jun 6, 2021
    Likes Received:
    1,692
    Trophy Points:
    93
    I don't mean this in a bad way, but please post less. You aren't adding anything to this specific conversation other than your beliefs, opinions, and conjectures. Everyone's got an idea of this or that. Everything you've said, everyone has heard here since 2011. Pointless to discuss or argue. Let's move on.

    I do appreciate questions or challenges to what I am doing here. Random thoughts on the world of measurements, not so much. Might be a good idea for you to review the content on the site, look at @atomicbob's work on amps and DACs or my work with headphones - which pushes past just frequency or steady-state sine wave distortion at 1kHz for measurements.
     
    Last edited: Jul 1, 2021
  16. Marvey

    Marvey Super Friend

    Staff Member Friend BWC
    Joined:
    Jun 6, 2021
    Likes Received:
    1,692
    Trophy Points:
    93
    Reality check time. I took these back to back. @Biodegraded: HD650 are mounted on the miniDSP EARS coupler. I'm also only taking results from the best channel (L) which didn't have the measurement defect.

    764 mVrms, 37Hz, Sennheiser HD650 as load. Unity gain.
    upload_2021-7-1_23-20-57.png
    upload_2021-7-1_22-55-11.png

    Dummy 300 ohm resistor load, again note results taken from the good L channel.
    upload_2021-7-1_23-23-0.png
    upload_2021-7-1_23-14-14.png
    Note: odd thing is that the left channel which as much less THD has higher H3.

    I am pretty certain of these results. The next thing is to compare to a solid-state amp with discrete parts.
     
    Last edited: Jul 1, 2021
  17. Beefy

    Beefy Almost "Made"

    Joined:
    Apr 10, 2021
    Likes Received:
    1,306
    Trophy Points:
    93
    Location:
    Canada
    Absolutely fascinating. Perhaps a high-z-out OTL tube amp with zero global feedback would give a more stark contrast?
     
  18. Marvey

    Marvey Super Friend

    Staff Member Friend BWC
    Joined:
    Jun 6, 2021
    Likes Received:
    1,692
    Trophy Points:
    93
    Not quite apples to apples because of high-gain (which typically results in higher distortion across the board). This is using the HD650 as the load.
    upload_2021-7-2_0-3-55.png
     
  19. Marvey

    Marvey Super Friend

    Staff Member Friend BWC
    Joined:
    Jun 6, 2021
    Likes Received:
    1,692
    Trophy Points:
    93
    Jotunheim 1. Low Gain. Sennheiser HD650 as load.
    upload_2021-7-2_0-19-20.png
    upload_2021-7-2_0-40-9.png
    Note: AC noise multiples at 60Hz, 120Hz, 180Hz, etc. Second harmonic of 37Hz is 74Hz.

    Back to Topping L30 for comparison:
    [​IMG]
    [​IMG]

    SINAD (1kHz at 2V into 300-ohm loads) can suck my balls.
    [​IMG]
     
    • Epic Epic x 6
    • Like Like x 5
    • List
    Last edited: Jul 2, 2021
  20. Nimrodor

    Nimrodor Rando

    Joined:
    Dec 26, 2017
    Likes Received:
    11
    Trophy Points:
    3
    Location:
    California
    I wonder if it could also be the capacitive load.

    Either way it might be worth checking if Magnius has similar behavior, since it also uses a composite TPA6120 output.
     

Share This Page