9 UGLY TRUTHS ABOUT HIGH END AUDIO - is there any more?

Measuring extremely well by studio standards is quite expensive because studios also look for consistency and reliability ($2-4k). Only the "we get excited by graphs" folks over at audiosciencereview seem to consider stuff made by dodgy manufacturers high end based on what, I suspect, are cherry picked samples

Exactly! China made products tend to have large variances. Batch to batch, sample to sample. The photos they post as marketing material they don't feel the urge to adhere to. In many cases, production version of the stuff doesn't use the same exact parts as their marketing material advertises.
 
Same feeling here.

what does he meant by high end dac? What’s his criteria?

Is it pricey and expensive ones?
Or is it MEASURING EXTREMELY WELL?
And why should a DAC add sonic signatures to the audio source? Isnt the purpose of a DAC - to reproduce as truthfully as it was originally recorded in the current digital format/file?
How can a DAC make a bad recording sound good…..if it remains a truthful DAC? (I mean it can’t add coloration to the source, right?)

I am beginning to think audiophiles will not consider a cheap DAC which measures extremely good - a high end DAC. :) The DAC has to have heavy esthetics and price .
A dac, Ofcourse , CAN make a recording sound different from how it was recorded. Good and bad is subjective. Qutest is known to smoothen out rough edges so considered a feel good dac. Why shd A DAC want to add color to original recording a philosophical and a pointless discussion because DAC do coloration. There is a market for it.

I agree with your other point on the deliberate mystifying and deifying of DACs even as they get more consistently to cd quality.
 
Exactly! China made products tend to have large variances. Batch to batch, sample to sample. The photos they post as marketing material they don't feel the urge to adhere to. In many cases, production version of the stuff doesn't use the same exact parts as their marketing material advertises.
Don't think it's necessarily about a country but about specific brands
 
Don't think it's necessarily about a country but about specific brands

Yes, I didn't articulate it that carefully.

Basically it is down to which factory the product in question is made, how is the QC there, and more importantly who defines/controls the QC. In the same country, products made for say HP/Apple/Dell/Samsung have excellent quality. So it is down to who is in the charge of QC.
 
Measuring extremely well by studio standards is quite expensive because studios also look for consistency and reliability ($2-4k). Only the "we get excited by graphs" folks over at audiosciencereview seem to consider stuff made by dodgy manufacturers high end based on what, I suspect, are cherry picked samples
Well I am also ‘folks over at audioscience’. I am not excited by the graphs…….but rather like the science and the scientific temper. (Btw I am not starting a tu-to-main-main of subjectivist vs objectivist).

I like well measured gadgets……including audio ones. In our field we have no room for errors/coloration. Hence we have something called as medical grade. AND THESE ARE BASED ON MEASUREMENTS AND EXPECTED PERFORMANCE & RELIABILITY.

Just maybe …..or even philosophically I just disagree that a DAC …which is a device to convert digital signals to audio should add coloration of its own. I am not listening to the DAC but to the music and recording….so it’s purpose is to convert back the signals to the original sounds. (Philosophically also I tend to go like this :) )
Well the cherry picking parts ……well there are not that too much evidences around. I mean my DENON was bricked by Sound United updates (without anything wrong done by me). So I guess some standard deviations will be applicable even for highly precision gadgets also.
The moot point is : how much value you are willing to give to a gadget and according to what your subjective/objective parameters? My subjective ones are not much reliable hence I bank on objectives…especially where engineering and electronics are involved. I mean I can see the chip industry as nothing but well measuring tech…the margin of error if large then we may not have computing.
 
Well I am also ‘folks over at audioscience’. I am not excited by the graphs…….but rather like the science and the scientific temper. (Btw I am not starting a tu-to-main-main of subjectivist vs objectivist).

I like well measured gadgets……including audio ones. In our field we have no room for errors/coloration. Hence we have something called as medical grade. AND THESE ARE BASED ON MEASUREMENTS AND EXPECTED PERFORMANCE & RELIABILITY.

Just maybe …..or even philosophically I just disagree that a DAC …which is a device to convert digital signals to audio should add coloration of its own. I am not listening to the DAC but to the music and recording….so it’s purpose is to convert back the signals to the original sounds. (Philosophically also I tend to go like this :) )
Well the cherry picking parts ……well there are not that too much evidences around. I mean my DENON was bricked by Sound United updates (without anything wrong done by me). So I guess some standard deviations will be applicable even for highly precision gadgets also.
The moot point is : how much value you are willing to give to a gadget and according to what your subjective/objective parameters? My subjective ones are not much reliable hence I bank on objectives…especially where engineering and electronics are involved. I mean I can see the chip industry as nothing but well measuring tech…the margin of error if large then we may not have computing.
Underlying Computing technology is precise but applications of computing can be imprecise... Audio is an application of compute. Like astrology app, think of audio as another app.
 
Well I am also ‘folks over at audioscience’. I am not excited by the graphs…….but rather like the science and the scientific temper. (Btw I am not starting a tu-to-main-main of subjectivist vs objectivist).

I like well measured gadgets……including audio ones. In our field we have no room for errors/coloration. Hence we have something called as medical grade. AND THESE ARE BASED ON MEASUREMENTS AND EXPECTED PERFORMANCE & RELIABILITY.
I get where you're coming from on well measured. I look for that as well but find ASR unreliable for that. Barring the owners pro harman group bias, let me give you another example of why. The science behind double blind tests shows people can pick up on differences in volumes of 0.2db. Extending that logic, any dac which errs +/-0.1db from flat in both directions is objectively not neutral. Yet these guys use a a +/-5db charts with 0.5db markings to show that everything is neutral. If they were true to the science, they would be showing +/-0.5db charts with 0.1db differences and highlighting diversions rather than giving way to dogma and saying everything sounds the same
 
Underlying Computing technology is precise but applications of computing can be imprecise... Audio is an application of compute. Like astrology app, think of audio as another app.
I respectfully disagree.
Audio reproduction is precise technology.

Music production is as you say like astrology. I like my good music …..to be reproduced by well engineered and measured audio devices.

I get where you're coming from on well measured. I look for that as well but find ASR unreliable for that. Barring the owners pro harman group bias, let me give you another example of why. The science behind double blind tests shows people can pick up on differences in volumes of 0.2db. Extending that logic, any dac which errs +/-0.1db from flat in both directions is objectively not neutral. Yet these guys use a a +/-5db charts with 0.5db markings to show that everything is neutral. If they were true to the science, they would be showing +/-0.5db charts with 0.1db differences and highlighting diversions rather than giving way to dogma and saying everything sounds the same
IMO - Amir uses an expensive analyser. And that audio analysers are also used by audio engineers.

And as far as I understood blind test are ….or should be done level matched. It is precisely because people can pick up 0.2db difference.
 
I respectfully disagree.
Audio reproduction is precise technology.

Music production is as you say like astrology. I like my good music …..to be reproduced by well engineered and measured au

I respectfully disagree.
Audio reproduction is precise technology.

Music production is as you say like astrology. I like my good music …..to be reproduced by well engineered and measured audio devices.
Audio reproduction done by me is my business... I choose how i like my reproduction, precise or colored or anything halfway is upto me. Music production Ofcourse is in artist domain and hence it's given that it's art not science.
 
IMO - Amir uses an expensive analyser. And that audio analysers are also used by audio engineers.

And as far as I understood blind test are ….or should be done level matched. It is precisely because people can pick up 0.2db difference.
Having a tool and knowing how to use it are two different things. The 0.2db data point was established through double blind tests. The fact that Amir seems to ignore it doesn't lend confidence
 
Having a tool and knowing how to use it are two different things. The 0.2db data point was established through double blind tests. The fact that Amir seems to ignore it doesn't lend confidence
I am getting confused here.

For me the charts show: the level of noise. And the SNR. And Amir and ASR members agree that we don’t need more than a SINAD of 115…..because human ears cannot resolve more than that. So I look for devices which have more SINAD …..for a SOTA experience for me (personally).

Audio reproduction done by me is my business... I choose how i like my reproduction, precise or colored or anything halfway is upto me. Music production Ofcourse is in artist domain and hence it's given that it's art not science.
I feel we are mixing something here.

when I say audio reproduction, I implied the electronics/engineering/and the amplification which are based on science.

But yes I think we both agree music production is art.

My submission is that audio should reproduce the music ….as originally made by the artist/composer- how he wanted to sound it. So transparency rather than coloration is what I look for in audio gadgets (and if they don’t measure well, how can they be transparent? )
 
I am getting confused here.

For me the charts show: the level of noise. And the SNR. And Amir and ASR members agree that we don’t need more than a SINAD of 115…..because human ears cannot resolve more than that. So I look for devices which have more SINAD …..for a SOTA experience for me (personally).
A question for you - what is the 115 threshold based on? Has the theory that people can't hear beyond that been tested through double blind tests? If no, does it still meet standards to be accepted as proven fact? Given you are an objectivist, I'm assuming you've investigated this rather than just accepting hearsay on an Internet forum

My earlier posts have been referring to linearity, where 0.2db is considered the audability threshold (read Peter Tooles research on this). This implies that if a piece of equipment varies more than 0.1db on both sides of a flat line, people would be expected to hear it in a double blind test. The frequency measurements the ASR folks publish don't have the resolution to see such fine divergences and they don't draw attention to them. Without addressing this, focusing on a single distortion and noise is much less relevant.

If you are interested in the science and are an objectivist, don't restrict yourself to the dogma of a single Internet forum. There is a world of research out there. The Harbeth forum is worth checking out, as is SBAF and the golden sound blog. I'm sure there are other sources as well that people can guide you to if you ask.
Edit: one point to add - the SBAF guys have their own biases (similar to ASR)
 
A question for you - what is the 115 threshold based on? Has the theory that people can't hear beyond that been tested through double blind tests? If no, does it still meet standards to be accepted as proven fact? Given you are an objectivist, I'm assuming you've investigated this rather than just accepting hearsay on an Internet forum

My earlier posts have been referring to linearity, where 0.2db is considered the audability threshold (read Peter Tooles research on this). This implies that if a piece of equipment varies more than 0.1db on both sides of a flat line, people would be expected to hear it in a double blind test. The frequency measurements the ASR folks publish don't have the resolution to see such fine divergences and they don't draw attention to them. Without addressing this, focusing on a single distortion and noise is much less relevant.

If you are interested in the science and are an objectivist, don't restrict yourself to the dogma of a single Internet forum. There is a world of research out there. The Harbeth forum is worth checking out, as is SBAF and the golden sound blog. I'm sure there are other sources as well that people can guide you to if you ask.
Edit: one point to add - the SBAF guys have their own biases (similar to ASR)
Thanks.


However I am not into that deep in audiophile realm.

It seems like we are going towards a debate. However therein my interest gets different - I am not audio expert, and its not my domain. I do have read few papers of Dr Toole (which btw is also a member of ASR and can be seen interactive ) only with a purpose of understanding whether what I buy as audio gadgets are good enough for me to listen to my music.

PS: Sharing some of my opinions here. I am not trying to win an argument. :) (I don't think it serves any purpose)
 
I do have read few papers of Dr Toole (which btw is also a member of ASR and can be seen interactive ) only with a purpose of understanding whether what I buy as audio gadgets are good enough for me to listen to my music.
I was misremembering the name yesterday. It's Floyd Toole, not Peter. Referring to the work he did before joining Harmann
 
Having a tool and knowing how to use it are two different things. The 0.2db data point was established through double blind tests. The fact that Amir seems to ignore it doesn't lend confidence
Confidence in what? His ratings of which dacs are better than others? I know some get miffed because their dac is rated noisy or not good.
 
I am getting confused here.

For me the charts show: the level of noise. And the SNR. And Amir and ASR members agree that we don’t need more than a SINAD of 115…..because human ears cannot resolve more than that. So I look for devices which have more SINAD …..for a SOTA experience for me (personally).


I feel we are mixing something here.

when I say audio reproduction, I implied the electronics/engineering/and the amplification which are based on science.

But yes I think we both agree music production is art.

My submission is that audio should reproduce the music ….as originally made by the artist/composer- how he wanted to sound it. So transparency rather than coloration is what I look for in audio gadgets (and if they don’t measure well, how can they be transparent? )
For simplification let's call what we do as designing music playback using discretes. The kind of dac, amp, eq, dap and player software available today allows me to playback music with my own coloration. Granted it takes times to figure out how and what impacts the sound...just the same way it's equally hard to get neutral presentation with discretes. It takes time and effort but it's an individual taste and people's taste differs and there are enough varieties available to cater to people's taste.
 
Colour of music is a very personal preference and Fletcher Munson curves with its later enhancements prove that each ear hears the same music differently.

Trying to hear it as the original artist/Sound music director intended is the nirvana most of us strive for , within limitations of rooms/content and budgets.

Of course..anyone can choose to ( and many do) enjoy music out of the standard earphones that one with phones itself. The idea of our hobby is to see how much more we can coax out of the music by means of components, acoustics and tweaks and thats why we visit forums ie to learn what more can be done to better it.
 
See post no 33 in this thread
OK just read it again. You have a problem with their methodology. I got that. But as a fairly new entrant to all this reading I can say they seem to be leading the discussion on measurements. That says something. The chances of their method being ultimately being universal is pretty high I would think. ASR maintains their publication in a truly catalogue form with easy to dive in literature and consistent test pattern. Its moderated and published reasonably in accordance with 1-2 page scientific paper. I have gone a little bit into other forums it's not as well centralized and documented. I see a difference there itself.

//Without addressing this, focusing on a single distortion and noise is much less relevant.//

Why the two cannot be separate and possibly be equally important? Do you mean to say distortion and noise measurements are not indicative of good equipment?
 
OK just read it again. You have a problem with their methodology. I got that. But as a fairly new entrant to all this reading I can say they seem to be leading the discussion on measurements. That says something. The chances of their method being ultimately being universal is pretty high I would think. ASR maintains their publication in a truly catalogue form with easy to dive in literature and consistent test pattern. Its moderated and published reasonably in accordance with 1-2 page scientific paper. I have gone a little bit into other forums it's not as well centralized and documented. I see a difference there itself.

//Without addressing this, focusing on a single distortion and noise is much less relevant.//

Why the two cannot be separate and possibly be equally important? Do you mean to say distortion and noise measurements are not indicative of good equipment?
I agree it’s a good resource. It’s just not the final authority on sound quality that it’s fans make it out to be. The same holds true for a range of audio review sites. The same issues that other review sites have about hazy commercial engagements and biases also apply here.

on the linearity vs noise question, the both could be equally important or, depending on the application, one could be more important than the other. Pretending that a single combined noise and distortion variable captures everything about measured sound quality is misleading
 
Join WhatsApp group to get HiFiMART.com Offers & Deals delivered to your smartphone!
Back
Top