I don't want to beat a dead horse but I'm bugged.


I just can't clear my head of this. I don't want to start a measurements vs listening war and I'd appreciate it if you guys don't, but I bought a Rogue Sphinx V3 as some of you may remember and have been enjoying it quite a bit. So, I head over to AVS and read Amir's review and he just rips it apart. But that's OK, measurements are measurements, that is not what bugs me. I learned in the early 70s that distortion numbers, etc, may not be that important to me. Then I read that he didn't even bother listening to the darn thing. That is what really bugs me. If something measures so poorly, wouldn't you want to correlate the measurements with what you hear? Do people still buy gear on measurements alone? I learned that can be a big mistake. I just don't get it, never have. Can anybody provide some insight to why some people are stuck on audio measurements? Help me package that so I can at least understand what they are thinking without dismissing them completely as a bunch of mislead sheep. 

russ69

@atmasphere When I'm speaking of accuracy and it's association with color, I'm thinking about two components that measure exactly the same in every measurement undertaken. Now lets say those measurements are such that one was deemed to be an accurate component. Will these  two components, or any number of components tested, meeting the above criteria necessarily sound exactly the same.

There is a failure of logic in this post. If all the proper measurements are taken and they are the same, the two pieces will sound the same too.

Since the rest of your post is based on a faulty premise, you might want to rethink this.

40 years ago it really wasn't practical to do the measurements that we can today. Sometime in that period we turned a corner. But the important specs don't show up on spec sheets for the most part (I do remember seeing an Adcom spec sheet clearly showing an increase in distortion above 3KHz, showing why the amp had brightness and harshness- even 20 years ago, this stuff was starting to show up). But audiophiles for the most part are living their lives as if the only measurements we can make were those of the 1980s and before. Back in those days the spec sheets were the Emperor's New Clothes- an amp that looked good on paper rarely sounded good as well.

This problem of audiophiles living their lives according to how things were 30-40 years ago has caused a lot of suffering (and to be clear, when people have made up stories about life and life does not agree with those stories, that is the source of all human suffering). Back then, if the manufacturer, distributor or dealer's lips were moving, he was lying and this has been a way of life for so long that we collectively no longer think about the fact that we are being lied to, we just know and accept that we will have to take the damn thing home and listen to it to know how it really sounds!

We are living in a transitional time where the measurement tech has caught up with our subjective experience- now we can measure things that we hear with excellent correlation. The problem now is actually seeing the specs we need to see and knowing what they are telling us.

I've outlined all this previously.

Rather the point is that measurements aren’t reliably predictive of what a given audio product will sound like. They do not replace the act of listening.

@charles1dad 's comment here puts this dilemma in a nutshell- although I've always appreciated his comments as being some of the more level headed seen on this forum, the simple fact is that we have arrived at a point where the quote above was true at one point but isn't any more (although his advice of simply listening to see if it works for you is quite valid). If we do all the measurements (including in the room itself) we can quite reliably predict how things will turn out. Honestly it appears to me that people don't do all the homework. For example, how many here have run pink noise through their system to see how it fares in the room? FWIW these days that is quite easy compared to only a few years ago- as they say, 'there's an app for that'.

Pink noise can show you room issues, breakups in speaker cones, all sort of pesky stuff! It won't show you brightness caused by distortion, but you can get that information from proper measurements of the electronics.

FWIW I fault the 'objectivist' camp as much as the 'subjectivists' in this debate. Many of them don't know about all the specs that make a difference and allow their expectation bias to color their perceptions. IOW guilty of exactly the same thing as the people they fault. That is no way to make progress! Put another way if one is in possession of the facts, then one knows there is no good reason for objectivists and subjectivists to be at odds.

 

@atmasphere FWIW I fault the 'objectivist' camp as much as the 'subjectivists' in this debate. Many of them don't know about all the specs that make a difference and allow their expectation bias to color their perceptions. IOW guilty of exactly the same thing as the people they fault. That is no way to make progress! Put another way if one is in possession of the facts, then one knows there is no good reason for objectivists and subjectivists to be at odds.

Ralph,  this paragraph  succinctly nails the point I've tried to explain with numerous posts on this and other threads. The high level complete and "meaningful "  type of measuring audio electronics you are describing aren't being utilized .

The current type of measurements as practiced over at ASR and similar sources just are not predictive of how a product will sound. This is patently obvious,  no correlation to subsequent sound quality at all.

  Ralph what you are referring to in terms of measurements is far removed from what we are presently being provided from these sites and their advocates. I would sincerely welcome legitimate and predictive measurements.  Until this happens I'll just listen. BTW Thanks for your kind comments. 

Charles 

 

The current type of measurements as practiced over at ASR and similar sources just are not predictive of how a product will sound. This is patently obvious,  no correlation to subsequent sound quality at all.

How can you possibly state this with such confidence without proving that you or anyone can reliably detect differences after 2 components have been tested by a sight like ASR or equivalent.

- THD+N from 100mW to max power at 20Hz, 500Hz, 1-5-10-15KHz.

- Power versus distortion single frequency from 10mW to max

- Frequency response at 4R, which would allow extraction of output resistance

- 32 tone inter modulation tests. This would represent real music.

- I saw a 2 ohm test on a recent amp from 50mW and up

ON DACs add:

- frequency response at various input sample rates and with the different filters the DAC offers.

- jitter test

- usually tests all the input types, but not consistent

 

I am aware of some videos highlighting some potential corner conditions (at least with DACs) that ASR does not test for, but which may also not be an issue with real music. This still brings me back to my first paragraph. With real music can you detect issues?

Ralph what you are referring to in terms of measurements is far removed from what we are presently being provided from these sites and their advocates. I would sincerely welcome legitimate and predictive measurements.

@charles1dad ASR is actually providing a lot of information, much of it quite useful. They don’t always hit the nail on the head (see the PSAudio regenerator review where Amirim pretty well missed the point). But they do graph distortion vs frequency and usually show the distortion spectra. I’d like to see the distortion spectra a -6dB of full power in addition to 1 Watt, and also see the distortion vs frequency at 1 Watt, -6db of full power and also full power. They could do all that if they caused their hands to move. So my opinion of them is higher than yours; and I think they can do better with little effort.

If you see distortion rising with frequency take that as a Bad Thing (likely bright and harsh, not what you like if my guess is right 😉). Your SETs don’t do that, neither do our OTLs or any zero feedback amplifier. Amps that run feedback need to not do that too. So right there you know something very useful that is routine on that site.

 

To me ASR does pretty good with DAC measurements and Speakers, especially since he got the Klippel.