Dilemma: Should I not believe any reviewers who talk about cables or just ignore that section of their review?

May 28, 2012 at 10:20 AM Post #706 of 1,790
Quote:
Magick Man, I have not heard electrostatic speakers, but electrostatic headphones have a different experience in soundstage (or what I call soundspace) than speakers at the same price level.
 
Likewise, I think custom IEM's have a very different experience in soundspace / imaging / layering (= detail in a sense) than an LCD-2 at the same price.
 
So what we have at the $1000 mark is speakers, STAX, LCD-2 and custom IEM's, all with a very different presentation.  If science can't show me these differences, then it needs to keep looking.  I won't accept CSD / SWR / IR / THD+N or anything like that until it's correlated to the listening experience / data.  I don't accept random pseudo-science theories from thin air, like pure square-waves from a Nintendo = imaging.  That, alon with impulse response, seem like okay intuive theories, however they don't have any scientific data.  So I'm not sure why people are complaining that I'm acting anti-science here.

My teachers used to always tell, making music is math and to make out anything out of data is require's some analysis. It is true that many got creative to achieve ground breaking headphones, but it is almost usually verified by measurements TOGETHER with listening test. Square wave or the other test looks at fidelity to the source in a electrical method, our brains ain't AP analysers and can't decipher the data the way a scope does. Don't believe in HAT models? Binaural would not exist without it(together with that superb imaging). I think the current HATs models and AP analysers are necessary and able to somewhat(not totally) correlate with listening data, or else why sound engineers choose them if its pseudo-science as you said? There may have been new perceptual models used my audio engineers to test products that I may not be aware of though it is possible that they are a lot more accurate in THD and IMD in relating results to subjective evaluation. Never hurt to have accurate gear though.
 
Have you read the story of the Yamaha NS-1? It's a great example of how a speaker that was unpopular with the consumer market back in the day managed to make its way to top studios, as it measured well in terms of group delay(phase). It was measured after quite some time since its inception into the speaker market. If you don't accept IR and SWR theories(since both are interrelated*), I guess stuff like NOS/SACD/192khz has no real case either, all seem like pseudo science after all. 
 
Also for the 44.1khz and 88.2khz paper, if you read the paper, DAC performance maybe playing a part(lack of preference for 44.1khz native vs 44.1khz downsampled and the troubling statistics analysis) 
 
As for titanium vs plastic, I think metals sound better too, plastics just don't have the stiffness of titanium. Keep in mind the natural frequency of metals are vastly different from plastic. I do like nice old paper cone drivers though. 
 
For speakers vs headphones vs IEMs, they all interact with the ear differently. Headphones are more head size and position dependent, a slight shift could really upset that nice tonal balance unlike IEMs, I don't really like something sticking in my ear though(I has no customs) I think though, given a good budget of 10k(or even just 5k or less) for either SR-009 system(or other flagship system) or a good pair of monitors, I will always head for the latter. Even with my low-mid fi system(the R2000Ts), they give serious detail when setup with proper EQ and good soundstage. Even something like my NAD system will give serious detail with proper acoustic treatment (I once pissed my family of clearing the living room to setup them up properly, since then I used them sparingly in my room lol)
 
Edit: Looking forward to anyone peer reviewing the paper though, since one other paper seems to have contradicting results in terms of high-res formats is concerned. 
 
May 28, 2012 at 12:08 PM Post #707 of 1,790
Quote:
 
Are you going out of your way to misinterpret what I'm saying?  You even responded to what I said about how to measure a headphone's effect on the perception of soundstage but now I'm supposed to be saying that all headphone sound the same.  What?  Seriously I don't know what else to say...
 

 
I just decided to stop saying anything.
 
se
 
May 28, 2012 at 12:23 PM Post #709 of 1,790
Quote:
Have you read the story of the Yamaha NS-1? It's a great example of how a speaker that was unpopular with the consumer market back in the day managed to make its way to top studios, as it measured well in terms of group delay(phase). It was measured after quite some time since its inception into the speaker market. If you don't accept IR and SWR theories(since both are interrelated*), I guess stuff like NOS/SACD/192khz has no real case either, all seem like pseudo science after all. 
 

 
Not trying to start an argument, but on the whole, most people (including recording engineers) think that the NS-1 sounds terrible.
They are really just a tool to help a mix engineer (not a recording engineer or mastering engineer) pull a mix together.
 
May 28, 2012 at 1:27 PM Post #710 of 1,790
Very subtle differences can be more important than they appear at first glance,.


There are huge issues in achieving great sound and there are minute ones. Too many audiophools focus on differences that only exist on paper and avoid dealing with the things that really do matter.

If you want really good sound, get really good speakers, set up your listening room well, and equalize to correct for response imbalances. Those are the things that really matter, but most people don't do that.
 
May 28, 2012 at 1:45 PM Post #711 of 1,790
So what we have at the $1000 mark is speakers, STAX, LCD-2 and custom IEM's, all with a very different presentation.  If science can't show me these differences, then it needs to keep looking.


You're making it way too complicated for yourself. You don't need science at all, just your ears.

Here's what you should do... Go to a chamber music concert. Sit fourth row center and pay attention to how it sounds. Afterwards, come home and put on a well recorded CD of a strong quartet. Listen to it on a good speaker setup, and with various headphones. Choose the one that sounds most like the live concert you attended.

I know what you'll end up choosing... the same thing engineers and sound mixers all use.
 
May 28, 2012 at 1:50 PM Post #712 of 1,790
Not trying to start an argument, but on the whole, most people (including recording engineers) think that the NS-1 sounds terrible.
They are really just a tool to help a mix engineer (not a recording engineer or mastering engineer) pull a mix together.


Every studio I've ever worked in used JBL studio monitors and they sounded fantastic when calibrated to the room.
 
May 28, 2012 at 2:06 PM Post #713 of 1,790
Originally Posted by firev1 /img/forum/go_quote.gif
 
My teachers used to always tell, making music is math

 
Not sure if you're joking.  So anyway, you mean like 16 year olds that skip every math class and become very successful / talented musicians? 
 
 
Originally Posted by firev1 /img/forum/go_quote.gif
 
It is true that many got creative to achieve ground breaking headphones, but it is almost usually verified by measurements TOGETHER with listening test.

 
That's exactly what I wrote?  "Stax and JH used measurements and data, yes... - Look, there's all kinds of data, it's not skin deep, and you need intuition and creativity in order to invent something to measure in the first place."
 
Originally Posted by firev1 /img/forum/go_quote.gif
 
Don't believe in HAT models? Binaural would not exist without it(together with that superb imaging). I think the current HATs models and AP analysers are necessary and able to somewhat(not totally) correlate with listening data, or else why sound engineers choose them if its pseudo-science as you said?

 
- I never said I don't believe in HAT models.
- binaural in HATS has nothing to do with the imaging I presented to you very clearly.
- I never said HAT models are pseudo-science.
- You haven't presented any correlation to listening data, I presented the FX500 versus FX700 CSD as an example, I also presented the Qualia 010, Shure SE535, Sony MDR-7550 and Audeze LCD-3 in post #653, I have several more and I've elucidated what I've called pseudo-science, you're twisting it to say I'm calling HATS models pseudo-science ー then why am I looking at sites like http://sonove.angry.jp all the time.
 
Originally Posted by firev1 /img/forum/go_quote.gif
 
There may have been new perceptual models used my audio engineers to test products that I may not be aware of though it is possible that they are a lot more accurate in THD and IMD in relating results to subjective evaluation.

 
That isn't scientific data.  I looked at the Harman study of relating a speaker with flat FR to speakers with deviated FR to teenagers from America and Japan and their listening impressions favoured the flat FR model.
 
That makes sense since flat FR is desirable, however the speakers were different models, what if the flat FR speaker just sounds better, like the STAX SR-009 does?  What if I showed the teenagers the Sony MDR-EX700 versus the EX1000?  We already have the listening data on this one actually.
 
19a4bd94_FR_MDR_EX1000vs700.gif

 
 
You missed the point on laptop speakers and TV speakers versus a car audio system with a quality sub-woofer and titanium tweeters.
 
If I equalized the car audio system to look shocking in FR, and equalized my laptop speakers to look as flat as possible... which one sounds better to you?
 
 
Originally Posted by firev1 /img/forum/go_quote.gif
 
Have you read the story of the Yamaha NS-1? It's a great example of how a speaker that was unpopular with the consumer market back in the day managed to make its way to top studios, as it measured well in terms of group delay(phase). It was measured after quite some time since its inception into the speaker market. If you don't accept IR and SWR theories(since both are interrelated*), I guess stuff like NOS/SACD/192khz has no real case either, all seem like pseudo science after all.

 
I accept IR and SWR, I don't accept the unverified theories surrounding them, until you can link to some evidence, that's called science, isn't it?
NOS/SACD/192kHz is for a different thread.
 
I accept that a transducer with a perfect SWR is truer to the source, and thus more accurate, however not sure exactly where the accuracy lies, for my intuitive / spatial conclusion on what square-waves should represent, and for data such as...
 
graphCompare.php
graphCompare.php

 
 
 
Originally Posted by firev1 /img/forum/go_quote.gif
 
Also for the 44.1khz and 88.2khz paper, if you read the paper, DAC performance maybe playing a part(lack of preference for 44.1khz native vs 44.1khz downsampled and the troubling statistics analysis)

 
Yes I read the analysis, not the entire paper yet, the statistics aren't perfect however...
 
"Collapsed results of all 16 subjects showed significant different for 88.2 vs native 44.1 Orchestral excerpt (p =.01)"
 
This study isn't perfect however it is at the very least more scientific and intact than the Meyer & Moran study, where I have read the entire paper and statistical post-analysis.
 
For example check post #139 here - http://www.sa-cd.net/showthread/42987//y?page=14, this thread is directly interactive with the author of the test.
 
The Meyer & Moran study uses flawed logic to 'prove' the transparency of the A/D/A (A/DC->D/AC) transfer as well.  If anything, if the A/D/A transfer is totally transparent, then they could have devised an A/D/A/D/A/D/A/D/A/D/A/D/A/D/A/D/A transfer with total transparency.
 
However... the results would still be subject to the transparency of the playback system (no transducer is totally transparent, they all instill their own sound into the final signal), the participants acuity, and the blind testing setup ー which I've covered earlier in this thread, with examples like the perfume switch... and... how to ABX fluroescent lamps?  How to ABX UVA/UVC light?  How to ABX a visual which changes consciousness?  How to ABX Tetris versus PTSD?  How to ABX audio/visual input latency?  How to find human reaction speed?  How to account for pattern identification? etc.
 
 
Originally Posted by firev1 /img/forum/go_quote.gif
 
As for titanium vs plastic, I think metals sound better too,

 
Then you've made a human listening evaluation, show me why titanium has more detail / velocity / shine whatever in the FR / SWR / IR, etc!
 
May 28, 2012 at 2:11 PM Post #714 of 1,790
Quote:
 
So what we have at the $1000 mark is speakers, STAX, LCD-2 and custom IEM's, all with a very different presentation.  If science can't show me these differences, then it needs to keep looking.  (Agree) I won't accept CSD / SWR / IR / THD+N or anything like that until it's correlated to the listening experience / data. (Agree with the caveat below)  I don't accept random pseudo-science theories from thin air, like pure square-waves from a Nintendo = imaging.  (Agree) That, alon with impulse response, seem like okay intuive theories, however they don't have any scientific data. (Agree - but if they mesh with predicted behavior, should be investigated not dismissed) So I'm not sure why people are complaining that I'm acting anti-science here.

 
I actually don't disagree with you in principle here. I agree, we need more data, and correlation to listener experience with the measurement data we do have will be important. However, we do not yet have that correlation (that I know of - certainly not demonstrated here - though we do have many good theories on how they are applied), and we do have ample data that listener experience is highly variable and subject to many biases and flaws from a scientific standpoint. 
 
So in the meantime - do we simply ignore the objective data we do have? Simply because we are not yet able to say with certainty what it means? That leaves us with nothing (user experience being too unreliable).
 
The scientific view, would be to keep collecting the data, and then develop a theory that explains the data we have (and offers predictive value) - then experiment to test the hypothesis and see. Not to walk away from it until it is 100% proven. 
 
We say you are being anti-science here because you are looking for an answer, not a process. Science is a process, and can offer insight and useful tools, even before it "knows" the answer you are looking for. 
 
May 28, 2012 at 3:06 PM Post #715 of 1,790
You're making it way too complicated for yourself. You don't need science at all, just your ears.

Here's what you should do... Go to a chamber music concert. Sit fourth row center and pay attention to how it sounds. Afterwards, come home and put on a well recorded CD of a strong quartet. Listen to it on a good speaker setup, and with various headphones. Choose the one that sounds most like the live concert you attended.

 
That's why I bought the Audio Technica A2000X because that's exactly what the listener did.  Your posts are all making sense to me however I like the differences which I can't find on paper and subtle differences which you whisk off, like information above 16kHz?  Plus, I like the sonics of IEM's (however feel free to recommend a pair of bookshelf speakers with pristine crystal clarity)...
 
The pseudo-science clientele are saying "paper first, truth later", clearly they are looking for honey in Alaska.
 
You need to experience audio for yourself and then pinpoint the experience on paper.  Maverickronin attested he has heard an inherent soundstage in speaker / HP / IEM's seperate from the audio signal, and devised a theory of reference speaker->HATS to possibly find it on paper in the future,
 
so... a vivid difference is okay for him, a subtle difference needs evidence first?
 
If there is no scientific evidence of something so colossal, no paper or correlation to listening anywhere, then how can audio science at the same time cover all the very slight differences?
 
?
 
May 28, 2012 at 3:38 PM Post #716 of 1,790
http://www.alaskabirchsyrup.com/rawalwiho1lb.html
 
Alaskan honey. Not sure how that came to be the analogy. 
 
May 28, 2012 at 6:27 PM Post #717 of 1,790
You need to experience audio for yourself and then pinpoint the experience on paper.


You've got that exactly backwards. The idea is to understand how sound works first, then do the listening to try to identify the things you've read about. If you had done that, you would know exactly what sound above 16kHz sounds like and just how subtle subtle is. If you listen first and try to chase down what you heard, you'll go down a rabbit hole of a dozen different explanations and not know which one is what you're actually hearing.
 
May 28, 2012 at 7:28 PM Post #719 of 1,790
Not test tones. Music. Listen to what sort of information there is above 16kHz in music. I think you'll be surprised.
 
May 28, 2012 at 7:56 PM Post #720 of 1,790
Yeah anyway I'll ABX it with music and I don't have it backwards, pseudo-science has it backwards, I listen first and go down the rabbit hole that's exactly how real science works.
 
I mean for example subliminal advertising existed long before it was proved, that's rabbit hole territory right?, and the pseudo-science audio avengers on their high horse of sine-waves and voltage levels would say "subliminal advertising is complete nonsense" <-- until you can ABX it. <-- which is impossible <-- lol.
 

Users who are viewing this thread

Back
Top