AVS Forum banner
1 - 14 of 14 Posts

·
Registered
Joined
·
16,705 Posts
Discussion Starter · #1 ·

·
Banned
Joined
·
14,420 Posts

Quote:
Originally Posted by cel4145  /t/1519888/toms-hardware-article-on-dacs-headphone-amp-output#post_24409527


Just came out today.

What Does It Take To Turn The PC Into A Hi-Fi Audio Platform?


More evidence that spending a fortune on DACs is a fools errand.

The big problem I see is that they did a questionable job of level matching, among other things.


A questionable test is a questionable test whether or not the results are like tests I've done myself or not.
 

·
Registered
Joined
·
2,354 Posts

Quote:
Originally Posted by arnyk  /t/1519888/toms-hardware-article-on-dacs-headphone-amp-output/0_100#post_24410703


The big problem I see is that they did a questionable job of level matching, among other things.


A questionable test is a questionable test whether or not the results are like tests I've done myself or not.

I take that you would rather see them measure voltage, at the headphone impedance, off the output?


I agree with them that the Benchmark is too low of a value proposition. The HDR1 didn't do any better or worse than my $170 EMU 1212M PCIe.
 

·
Registered
Joined
·
16,705 Posts
Discussion Starter · #4 ·

Quote:
Originally Posted by arnyk  /t/1519888/toms-hardware-article-on-dacs-headphone-amp-output#post_24410703


The big problem I see is that they did a questionable job of level matching, among other things.


A questionable test is a questionable test whether or not the results are like tests I've done myself or not.

They acknowledged that the level matching was not perfect and explained why.


Besides, you are evaluating the methodology out of context. Their primary claim in their conclusion is that the Realtek offers excellent sound quality to the two listeners in comparison to the other hardware, and more refined volume leveling would not offer any more information for how that would affect real world use. Now if they were trying to claim significant differences, you would be correct. As it is, your point is moot.
 

·
Banned
Joined
·
14,420 Posts

Quote:
Originally Posted by Jinjuku  /t/1519888/toms-hardware-article-on-dacs-headphone-amp-output#post_24410790

Quote:
Originally Posted by arnyk  /t/1519888/toms-hardware-article-on-dacs-headphone-amp-output/0_100#post_24410703


The big problem I see is that they did a questionable job of level matching, among other things.


A questionable test is a questionable test whether or not the results are like tests I've done myself or not.

I take that you would rather see them measure voltage, at the headphone impedance, off the output?

Very close if not right on - I'd like to see them measure the voltage across the headphone jack. It is very easy to make up a little cable to do this.


These guys brag themselves up as great experimentalists and then get tripped up by a nit.


(head shaking)
 

·
Banned
Joined
·
14,420 Posts

Quote:
Originally Posted by cel4145  /t/1519888/toms-hardware-article-on-dacs-headphone-amp-output#post_24410838

Quote:
Originally Posted by arnyk  /t/1519888/toms-hardware-article-on-dacs-headphone-amp-output#post_24410703


The big problem I see is that they did a questionable job of level matching, among other things.


A questionable test is a questionable test whether or not the results are like tests I've done myself or not.

They acknowledged that the level matching was not perfect and explained why.


Besides, you are evaluating the methodology out of context. Their primary claim in their conclusion is that the Realtek offers excellent sound quality to the two listeners in comparison to the other hardware, and more refined volume leveling would not offer any more information for how that would affect real world use. Now if they were trying to claim significant differences, you would be correct. As it is, your point is moot.

Here is a counter argument:


The level mismatch distracted them from hearing differences that they would have otherwise noticed and reported.
 

·
Registered
Joined
·
2,354 Posts

Quote:
Originally Posted by arnyk  /t/1519888/toms-hardware-article-on-dacs-headphone-amp-output/0_100#post_24410893


Here is a counter argument:


The level mismatch distracted them from hearing differences that they would have otherwise noticed and reported.

A level mismatch would be interesting if it was done purposefully (1.5 dB hotter or cooler) as control variable.
 

·
Registered
Joined
·
16,705 Posts
Discussion Starter · #8 ·

Quote:
Originally Posted by arnyk  /t/1519888/toms-hardware-article-on-dacs-headphone-amp-output#post_24410893


Here is a counter argument:


The level mismatch distracted them from hearing differences that they would have otherwise noticed and reported.

Still moot from my perspective. For real world use, those very minor differences would be insignificant. If one has to listen to it more closely matched than their test setup, then it doesn't really matter to support their claim that the Realtec has "Great sound quality."


Now I could agree that maybe for audio engineers doing production work the difference could be important, but that's not their audience or what they are claiming. I suspect you are biased against the test because it doesn't meet your standards for testing methodology purity (or for some other reason), and thus are not evaluating the methodology properly in relation to the claims. Then again, my bias is that I'm not a pure skeptic. I believe in the importance of the role of the believing game in critical thinking . IMO, pure skepticism creates a blind spot necessary for robust critical analysis.
 

·
Banned
Joined
·
18,829 Posts

Quote:
Originally Posted by arnyk  /t/1519888/toms-hardware-article-on-dacs-headphone-amp-output#post_24410888


Very close if not right on - I'd like to see them measure the voltage across the headphone jack. It is very easy to make up a little cable to do this.
I am not following you Arny. Why would you level match at that point?
 

·
Banned
Joined
·
14,420 Posts

Quote:
Originally Posted by cel4145  /t/1519888/toms-hardware-article-on-dacs-headphone-amp-output#post_24411037

Quote:
Originally Posted by arnyk  /t/1519888/toms-hardware-article-on-dacs-headphone-amp-output#post_24410893


Here is a counter argument:


The level mismatch distracted them from hearing differences that they would have otherwise noticed and reported.

I suspect you are biased against the test because it doesn't meet your standards for testing methodology purity (or for some other reason), and thus are not evaluating the methodology properly in relation to the claims.

Suspect no more, but let me rephrase.


The test does not meet my standards for accuracy of experimental controls. These standards are based on over 30 years of personal testing experience and working with some of the best and best known people in the audio business.


Remember, I invented ABX as applied to audio gear
 

·
Banned
Joined
·
14,420 Posts

Quote:
Originally Posted by amirm  /t/1519888/toms-hardware-article-on-dacs-headphone-amp-output#post_24411288

Quote:
Originally Posted by arnyk  /t/1519888/toms-hardware-article-on-dacs-headphone-amp-output#post_24410888


Very close if not right on - I'd like to see them measure the voltage across the headphone jack. It is very easy to make up a little cable to do this.
I am not following you Arny. Why would you level match at that point?

It is as accurate and reliable as any, and fast and easy.


Why not?
 

·
Banned
Joined
·
14,420 Posts

Quote:
Originally Posted by cel4145  /t/1519888/toms-hardware-article-on-dacs-headphone-amp-output#post_24411037

Quote:
Originally Posted by arnyk  /t/1519888/toms-hardware-article-on-dacs-headphone-amp-output#post_24410893


Here is a counter argument:


The level mismatch distracted them from hearing differences that they would have otherwise noticed and reported.

Still moot from my perspective. For real world use, those very minor differences would be insignificant. If one has to listen to it more closely matched than their test setup, then it doesn't really matter to support their claim that the Realtec has "Great sound quality."

BTW I do need to step off my ABX high horse long enough to say that from a practical standpoint I pretty much agree with you.


Reality is that ABX can sometimes be faulted for being unrealistically sensitive to audible differences that really don't matter.


A practical example of this can be seen right here on AVS. As many know I frequently tell people to bump their AVR digitally-controlled volume controls up or down a few notches and see what they think about it. AFAIK nobody has ever come back and said: "Yes I raised (or lowered) my volume control 3 dB (6 steps) and the difference was so pleasing and compelling to me that I ran right out and bought a..." And that is a sighted evaluation but it holds far more influences constant than your average audiophile stereo shop crawl listening evaluation.


If so many people weren't using sighted evaluations as their reference standard and can't see all of the false positives because of their prejudices, they might agree with this.
 

·
Banned
Joined
·
14,420 Posts

Quote:
Originally Posted by Jinjuku  /t/1519888/toms-hardware-article-on-dacs-headphone-amp-output#post_24411030

Quote:
Originally Posted by arnyk  /t/1519888/toms-hardware-article-on-dacs-headphone-amp-output/0_100#post_24410893


Here is a counter argument:


The level mismatch distracted them from hearing differences that they would have otherwise noticed and reported.

A level mismatch would be interesting if it was done purposefully (1.5 dB hotter or cooler) as control variable.

Further analysis shows that 4 out of the 8 devices tested have such high output impedance that most headphones that might be used to test them affect their frequency response enough to sound different.


For that reason, they are also likely to sound different from other products that actually have superior performance.


This issue is well known among techs who work with portable digital players and a high proportion of all modern digital players have effectively addressed the issue.


IOW there are $29 portable digital music players that drive headphones more accurately than about half the units tested. :-(


IOW they are something like SET amp that sound different with every different speaker you attach to them.
 

·
Registered
Joined
·
16,705 Posts
Discussion Starter · #14 ·

Quote:
Originally Posted by arnyk  /t/1519888/toms-hardware-article-on-dacs-headphone-amp-output#post_24411569


BTW I do need to step off my ABX high horse long enough to say that from a practical standpoint I pretty much agree with you.


Reality is that ABX can sometimes be faulted for being unrealistically sensitive to audible differences that really don't matter.


A practical example of this can be seen right here on AVS. As many know I frequently tell people to bump their AVR digitally-controlled volume controls up or down a few notches and see what they think about it. AFAIK nobody has ever come back and said: "Yes I raised (or lowered) my volume control 3 dB (6 steps) and the difference was so pleasing and compelling to me that I ran right out and bought a..." And that is a sighted evaluation but it holds far more influences constant than your average audiophile stereo shop crawl listening evaluation.


If so many people weren't using sighted evaluations as their reference standard and can't see all of the false positives because of their prejudices, they might agree with this.

I do not disagree at all that your experimental testing design methods are much more discriminating. I just believe that given the very, very general nature of their claim, that the methodology does work to make their claim. So is your methodology a better testing method overall as testing methods go? Sure. Does it matter here? I just don't think so. Sometimes more discrimination doesn't provide any more validity to a claim. At a certain point, there is enough evidence.


Now if they were trying to tease out relative differences between them, then I totally agree with everything you have said. But the author rated all four setups as "-Great sound quality." That's such a general subjective definition anyway.


In fact, they try to make the point in their conclusion that more discrimination really shouldn't matter in practical use. So at their level of volume leveling discrimination, it really wouldn't matter to me if more precise volume leveling demonstrated some fine difference. Everytime I put on my headphones, the slight variation in placement each time could make more difference in the sound than that. Ultimately it comes down to this: if two dac/headphone amps sound equivalent in SQ when one is run .5 db higher than the other, even though they sound slightly different when better level matched, that level of discrimination of difference is so fine, most people should not care or focus any attention on it. (lol)
 
1 - 14 of 14 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top