1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.

    Dismiss Notice

Objectivists board room

Discussion in 'Sound Science' started by joe bloggs, May 28, 2015.
First
 
Back
248 249 250 251 252 253 254 255 256 257
259 260 261 262 263 264 265 266 267 268
Next
 
Last
  1. pinnahertz
    I'm confused about your objectives. Are you trying to ascertain that the cables do or do not sound different, or prove that your test subjects can't tell cables apart? Those are actually very different objectives. You should always clearly and simply define the objective, preferably a single one.

    You won't get good resolution on the comparisons because the transition takes far too long. As has already been mentioned, auditory memory is very, very short. For best comparison results the transition time should be no more than a few milliseconds. As you can see, that would required construction of some rather specialized equipment.

    With you handling the switching and the cable choice with full knowledge of what they are, there is a serious lack of control, and it certainly won't be a true blind test. You will be introducing a massive amount of bias...your own...consciously or not. A true ABX/DBT eliminates the knowledge of the X choice from everyone involved until the data is collected and compiled.

    You no doubt fully anticipate the test subjects will "fail" the test, and that in itself is a bias. A true ABX/DBT has no pass/fail, it just generates subjective data. Every trial produces a good data point, so there is no "fail" as such.

    The test may be fun, but will technically be inconclusive because of the lack of controls, presence of biases, and massively long switch time. My guess is you'll receive objections and challenges to your test.
     
    artnoi and CarlosUnchained like this.
  2. artnoi
    Ok, I'll think about tweaking my test then. Thanks for advice! My objective is to conclude whether the subjects hear the differences that they claim to hear or not.
     
  3. Argyris Contributor
    I'll save you the trouble: no, they don't.

    The difficult task is convincing them that they don't, if they desperately want to believe otherwise. It doesn't matter how thorough your test or what methodology you employ, they'll find fault with it and claim that this invalidates the result.

    The key here is not to attack the claim that there's any difference, no matter how small; it's, as bigshot said earlier, to attack the inconsistency inherent in claiming that A) something makes a trivially identifiable difference, then turning around and using the excuse that B) the difference is too subtle to be revealed by testing. If A is true, than B cannot be--an effect that is obvious cannot mysteriously vanish whenever it's tested for.

    At least unless the audiophile cables and equipment are sentient and are messing with us for some reason, in which case I have even more reason not to buy any of it.
     
  4. Strangelove424
    After farting around with EQ for a long time, a little here, and a little there, I found a setting for both my headphones (Beyer DT880 and Sennheiser HD600) that makes them sound almost exactly the same, minus low end extension deficiencies in the HD600 that I can't really make up for with boost. It's very hard to tell them apart, I started to get confused about which I was wearing. The only major difference became the ergonomic design, you can tell the Sennhesiers because of the height and clamp. Makes me wonder if I really need more than one headphone if they are both technically competent enough to adapt to EQing.

    What really made the sudden focus apparent was Natalie Cole's Unforgettable duet with father Nat. What a talented family. My EQ settings finally did both of them justice, capturing the femininity of Natalie's articulations while still capturing Nat's burly heft.
     
  5. bigshot
    BURLY HEFT! That would make a great name for an actor.
     
  6. Strangelove424
    I'm afraid to ask what kind of actor you're thinking of.
     
  7. bigshot
    [​IMG]
     
  8. Strangelove424
    ok, ok... good 'ole American stage actor...

    whew
     
  9. Argyris Contributor
    When my DT880 was my main driver, I had excellent results EQ'ing it. I did various things and kept tweaking it over time, but the one constant I always came back to was two notches, one at ~6 kHz and one at ~8.5 kHz. At this point it doesn't quite sound like the HD 600 (the DT880 lacks the upper midrange lift and has too much upper treble), but the main trouble spots are ironed out, and on its own it sounds excellent. Since the DT880 has a well-controlled response without ringing or audible distortion, it's an excellent candidate for EQ. As far as technical ability, I honestly don't see a whole lot to separate the DT880 and HD 600. They're just tuned differently.
     
  10. Strangelove424
    Yes, one of the realizations I had while EQing treble on the DT880 was that I needed two dips! For me, it was at ~2-3khz and ~8-10khz. On the HD600's I need a dip at ~2-3khz too, so it might just be my sensitivity in that particular range. Recently, I came to the conclusion that the DT880 needed a slight 1-2db boost in the lower mids too. Once I did that, suddenly everything jelled and the upper mids blended into treble a lot better. I very much agree that the DT880 is an excellent candidate for EQing. I like to think of them as EQ chameleons. Compared with the HD600, the DT880 is more responsive to EQing, with more low end extension, but they are both phenomenally competent headphones. The Sennheiser perhaps needs less EQ to begin with, and has a more natural tuning straight from the factory.
     
  11. bigshot
    In my speaker system, I have a -4 dB notch with a very narrow Q at right about 8kHz. I kept hearing a distortion that sounded kind of like MP3 compression in the treble. It was like a pebble in my shoe- only in one tiny frequency spot. I tried to see where it was coming from thinking it might be a blown voice coil. But it came from all the speakers. So shrugged my shoulders and I notched it out. A year later, I found a review and bench test on my AVR that said that it had very low distortion, with the exception of a tiny little spike right at 8kHz. The reviewer guessed that it had something to do with the design of the amp and that it was probably inaudible because it was so narrow. I never would have guessed that something like that came from my amp and not my speakers.
     
  12. pinnahertz
    That can't be right. There is no distortion mechanism that would generate a fixed 'spike' independent of the signal. Distortion products are always related to the input signal in some way, and since that's changing, so would the distortion products. That 'spike' had to be some sort of noise, like from a switching regulator or something, or it wouldn't just be at 8k. Not to say your notch didn't solve a problem, though. It just likely wasn't distortion.
     
  13. bigshot
    Well, the guy running the test said it was a narrow band of distortion right at 8kHz. I just know that it irritated me. It was small enough that just dipping it out with a narrow EQ notch got rid of it.
     
  14. castleofargh Contributor
    not audio but once again demonstrating my testing skills. because it's important for people here to know how clever and professional I am, so that I can be the pillar everybody needs:
    I couldn't make sense of fluctuations in temperature that I didn't feel going that way in the room. because I trust most things before my senses, I didn't pay too much attention at first, but with the heat waves last month I had to accept that my feelings and temperature readings were in obvious conflict at times. sometimes 30° would feel fine, other times I would start melting gently. I checked humidity because I know it can clearly affect perceive temperature, but over a few weeks it became clear that it wasn't the only cause to me sometimes feeling fresher than what I would expect from reading the thermometer.
    there's a button that triggers a color code for a few seconds about weather prediction(rain and crap), I press it maybe twice a day(when going out). this same button also changes the backlight intensity in cycle(off/on/dimmed). turns out, light can come with heat! who knew except everybody? that extra heat apparently reaches the internal sensor if I leave the backlight ON.
    [​IMG]
    with the backlight ON I'm reading about 2 to 4°c more than with nothing at all right now.
    by then I felt pretty dumb, but it was nothing compared to how I felt when I remembered I had a little laser gun to measure the surface temperature and 2 other thermometers in the house all along. I never thought about using them to control the readings.


    anybody in need of professional testing, you can contact me at wwcastleofargh
     
  15. Arpiben
    If your AVR input signal is USB 2.0 you may have 8 kHz spurious spikes (as well as odd harmonics) not properly rejected.
    USB 2.0 (audio isochronous transfer) ->1-3 packet bursts every 0.125us (8kHz).
    Many other possibilities also.
     
First
 
Back
248 249 250 251 252 253 254 255 256 257
259 260 261 262 263 264 265 266 267 268
Next
 
Last

Share This Page