Any DAC above $2 is a scam you DO NOT pay for Audio Quality you pay for Features (Article)

[krabapple]

If the onboard DAC is that good, how come I can hear more dynamic and clear sound after changing from ALC892 to cheap SB, using the same speaker? The speaker was just a BeoPlay A6 which is hardly audiophile, and the main difference is not bass but overall. Are some shortcomings amplified by speakers? However I couldn't tell the difference years ago with a pair of Castle bookshelf which sounded softer and less dynamic, with another set of onboard and SB X-Fi Go.

Surely you can measure such obvious difference?

 

[nqa3]

Surely you can measure such obvious difference?

What measurement(s) should I look into? I used REW to measure frequency response - is it related?

BeoPlay was donated already; not sure if the ncore amp which I use currently could tell the difference yet, nothing about input on the spec. I do recall volume on BeoPlay had to be quiet high for line-in, near the level when distortion starts (only line-in, not other inputs).

 

[¥€$]

But I also pay for reliability, with the assumption that the DAC will last a lifetime without blown capacitors or fuses.

 

[phoenixdogfan]

I really liked this article by Toms and it sums it up perfectly, they had a bunch of senior audiophiles in this blind test and none of them were able to tell which was the $2 DAC and the $2000 DAC

https://www.tomshardware.com/reviews/high-end-pc-audio,3733-19.html

I constantly have to remind fanatic money wasting clowns on head fi about this all the time. Also burn in isn't real it is your brain adjusting to the sound, I have no idea why seemingly educated continue to spread propaganda and audio myths in the audiophile community. (well I do it has to do with buyers remorse)

Look how I got a HD 58x and the difference between the jubilee and my old HD 555 is small. I somewhat attributed it to my Xonar DG sound card with DAC and AMP built in not being up to scratch but lets be real that really isn't the case as my built in Onboard Asus Crystal Sound 2 audio is only a small downgrade from the Xonar DG card. Sure its noticeable but not a big amount. And I Want to bet alot of it can be fixed with an EQ

It's a classic article.

 

[phoenixdogfan]

I’ve owned DACs well above the $500-1000 range and I most certainly have moved in the opposite direction. I sold my “audiophile approved” DACs and am very happy with my Topping D10 and a heap of change to spend on music and wine (not necessarily in that order)

See, I woulda spent it on hookers and weed.

 

[Sir Sanders Zingmore]

See, I woulda spent it on hookers and weed.

reminds me of one of my favourite "Spinal Tap" quotes

• Marty DiBergi : What would you do if you couldn't play music anymore?
  Mick Shrimpton : Well, as long as there's, y'know, sex and drugs, I could do without the rock & roll.

 

[Abe_W]

Oh boy.....lol. hope they enjoy their realtek!

Realtek wielding clowns bwaaahaha...Jesus Christ,,,, even a basic utilitarian JoeSchmoe at my work who could give a ratsass about music in his everyday life can tell the difference between a realtek in a pc's motherboard and a topping d90 i keep at my work office these days. It seems to me that many of these audioscientists are just extreme skinflints or just genuinely poor as fk (my sympathies though if it were the latter).

When things are out of their financial reach, they become extremely disgruntled and start lying/calling bullsht on everything they can't afford. It is a personality trait really. I sense that ASR has a larger proportion of these dudes than some other forums out there. The test would be to give em something good for dirt cheap! I bet you they'll take it even though it was not "supposed" to sound any better than their realtek! All the froth mouthed dishonest bashing might disappear quietly! And then they'll start calling bullsht on the next pricier thing from that level onwards.

A polygraphed test should be conducted where the price of 10 dacs is the same. A polygraph is extremely critical to catch a constantly lying honors phd audioscientist. Realtek =10 dollars, Pyle =10 dollars, Schiit = 10 dollars, topping e30 = 10 dollars, Marantz HD Dac1 =10 dollars, Topping D90=10 dollars, Denafrips Venus =10 dollars, RME ADi = 10 dollars..etc. When everything is priced the same on a test, i bet you we'll see a whole new set of results "suddenly" about which DAC sounds better from audioscientist honors PhDs!

A polygraphed test with "affordability" removed from the equation is the only pathway to truth when it comes to audioscientists!

 

[BDWoody]

Realtek wielding clowns bwaaahaha...Jesus Christ,,,, even a basic utilitarian JoeSchmoe at my work who could give a ratsass about music in his everyday life can tell the difference between a realtek in a pc's motherboard and a topping d90 i keep at my work office these days. It seems to me that many of these audioscientists are just extreme skinflints or just genuinely poor as fk (my sympathies though if it were the latter).

When things are out of their financial reach, they become extremely disgruntled and start lying/calling bullsht on everything they can't afford. It is a personality trait really. I sense that ASR has a larger proportion of these dudes than some other forums out there. The test would be to give em something good for dirt cheap! I bet you they'll take it even though it was not "supposed" to sound any better than their realtek! All the froth mouthed dishonest bashing might disappear quietly! And then they'll start calling bullsht on the next pricier thing from that level onwards.

A polygraphed test should be conducted where the price of 10 dacs is the same. A polygraph is extremely critical to catch a constantly lying honors phd audioscientist. Realtek =10 dollars, Pyle =10 dollars, Schiit = 10 dollars, topping e30 = 10 dollars, Marantz HD Dac1 =10 dollars, Topping D90=10 dollars, Denafrips Venus =10 dollars, RME ADi = 10 dollars..etc. When everything is priced the same on a test, i bet you we'll see a whole new set of results "suddenly" about which DAC sounds better from audioscientist honors PhDs!

A polygraphed test with "affordability" removed from the equation is the only pathway to truth when it comes to audioscientists!

What is it with the trolls coming out of the woodwork of late? The quantity increases while the quality of the troll has gone way down.

It's a damn shame the state of trolling these days...

 

[ajawamnet]

I disagree...
1. most people could give a rats ass about fidelity
2. in blind test, humans have notoriously failed at discerning differences in hardware that's not obvious.

As I mention in Marketturd, the father of the opamp cookbooks - Walter Jung - was said to be able to hear the difference in 10 out 10 of a ceramic opamp vs. a epoxy packaged opamp. I was told this by one of the first FAE's at Linear Technology - Alan Rich.

As to me, well I design this shit: here's my lab:
http://www.ajawamnet.com

 

[ajawamnet]

As to BDWoody's troll comment - Awesome. Here's a vid that explains it:

 

[krabapple]

Realtek wielding clowns bwaaahaha...Jesus Christ,,,, even a basic utilitarian JoeSchmoe at my work who could give a ratsass about music in his everyday life can tell the difference between a realtek in a pc's motherboard and a topping d90 i keep at my work office these days. It seems to me that many of these audioscientists are just extreme skinflints or just genuinely poor as fk (my sympathies though if it were the latter).

Sure, go ahead set up a double blind, level matched comparison with some Joe Schmoes, and see how it works for you.

When things are out of their financial reach, they become extremely disgruntled and start lying/calling bullsht on everything they can't afford. It is a personality trait really. I sense that ASR has a larger proportion of these dudes than some other forums out there. The test would be to give em something good for dirt cheap! I bet you they'll take it even though it was not "supposed" to sound any better than their realtek! All the froth mouthed dishonest bashing might disappear quietly! And then they'll start calling bullsht on the next pricier thing from that level onwards.

I can afford your toys. I don't covet them. I sincerely believe your claims are foolish.

IOW, I'm amused by you and your ilk, not jealous.

A polygraphed test should be conducted where the price of 10 dacs is the same. A polygraph is extremely critical to catch a constantly lying honors phd audioscientist. Realtek =10 dollars, Pyle =10 dollars, Schiit = 10 dollars, topping e30 = 10 dollars, Marantz HD Dac1 =10 dollars, Topping D90=10 dollars, Denafrips Venus =10 dollars, RME ADi = 10 dollars..etc. When everything is priced the same on a test, i bet you we'll see a whole new set of results "suddenly" about which DAC sounds better from audioscientist honors PhDs!

Mmm, no, since a proper test would be double-blind, not 'polygraph' (which tests whether a person is lying...with only 87% accuracy even when analyzed by a trained 'reader')

Your repeated strange reference to 'audioscientist honors PhDs' makes you appear a bit unhinged, btw...or at least, to have an inferiority complex about people more intensively educated than you.

A polygraphed test with "affordability" removed from the equation is the only pathway to truth when it comes to audioscientists!

You clearly aren't enough of an authority on testing to make such claims.

 

[krabapple]

I disagree...
1. most people could give a rats ass about fidelity
2. in blind test, humans have notoriously failed at discerning differences in hardware that's not obvious.

Notoriously? Really? Better tell all lab-trained psychosensory researchers since about the 1950s then. As well as the journals they publish in, which all require blind protocols. This is big news. (Also, some evidence would be nice.)

Alas, in fact you have it quite backwards. It's sighted tests that are *notoriously* prone to error. Which is why good science in this area requires blind protocols.

As I mention in Marketturd, the father of the opamp cookbooks - Walter Jung - was said to be able to hear the difference in 10 out 10 of a ceramic opamp vs. a epoxy packaged opamp. I was told this by one of the first FAE's at Linear Technology - Alan Rich.

It's a lovely anecdote, thanks.

<pics with lots of screens>

Good luck with your hobby. Any books on sensory testing in there?

 

[carnagymelton]

I really liked this article by Toms and it sums it up perfectly, they had a bunch of senior audiophiles in this blind test and none of them were able to tell which was the $2 DAC and the $2000 DAC

https://www.tomshardware.com/reviews/high-end-pc-audio,3733-19.html

I constantly have to remind fanatic money wasting clowns on head fi about this all the time. Also burn in isn't real it is your brain adjusting to the sound, I have no idea why seemingly educated continue to spread propaganda and audio myths in the audiophile community. (well I do it has to do with buyers remorse)

Look how I got a HD 58x and the difference between the jubilee and my old HD 555 is small. I somewhat attributed it to my Xonar DG sound card with DAC and AMP built in not being up to scratch but lets be real that really isn't the case as my built in Onboard Asus Crystal Sound 2 audio is only a small downgrade from the Xonar DG card. Sure its noticeable but not a big amount. And I Want to bet alot of it can be fixed with an EQ

All of these tests have good intentions, but the complexity ruins the results. Why involve 5 DACs and ask people to identify each one? What good is that? That's nonsense. All one needs to do is take a $100 transparent DAC and compare it to an expensive "hi-fi" DAC (perhaps the Benchmark, or more expensive). KEEP IT SIMPLE. IF ABX test subjects cannot tell the difference between those DACs, THEN we are much closer to proving that a $100 transparent DAC is indistinguishable from a $5000 transparent DAC. I've seen many studies such as Tom's Hardware and the fact is that these studies cannot get out of their own way. They often involve 4 or more DACs, very confusing to listeners. Keep it simple, compare 2 DACS only!

 

[carnagymelton]

All of these tests have good intentions, but the complexity ruins the results. Why involve 5 DACs and ask people to identify each one? What good is that? That's nonsense. All one needs to do is take a $100 transparent DAC and compare it to an expensive "hi-fi" DAC (perhaps the Benchmark, or more expensive). KEEP IT SIMPLE. IF ABX test subjects cannot tell the difference between those DACs, THEN we are much closer to proving that a $100 transparent DAC is indistinguishable from a $5000 transparent DAC. I've seen many studies such as Tom's Hardware and the fact is that these studies cannot get out of their own way. They often involve 4 or more DACs, very confusing to listeners. Keep it simple, compare 2 DACS only!

Or perhaps the entire point of these tests IS confusion. Muddying up things so much with choices that the test subjects don't have a chance in health class of getting the right answer? At best, unnecessary complexity.

 

[ajawamnet]

Notoriously? Really? Better tell all lab-trained psychosensory researchers since about the 1950s then. As well as the journals they publish in, which all require blind protocols. This is big news. (Also, some evidence would be nice.)

Alas, in fact you have it quite backwards. It's sighted tests that are *notoriously* prone to error. Which is why good science in this area requires blind protocols.




It's a lovely anecdote, thanks.

<pics with lots of screens>

Good luck with your hobby. Any books on sensory testing in there?

Not a hobby , I've done over 3,000 designs in 40 years - 300 with the guy that was the president of Vega Wireless I hold 3 patents so far including one that looks very similar to the internet of things http://www.ajawamnet.com/amnet/

As to sensory testing - I'll let Prof Land discuss human hearing and testing - it gets a bit deep- it is a college course from Cornell:

Note where he shows the variance at about 3:37 in regards to just localization.

Now compound that with the variability of each individual with regard to perception of spectral content, which usually manifests itself in perceived distortion, each human will NOT aurally sense in the same manner.

And if you're really bored you can watch this on human hearing, how frail it is, again how localization works - again an MIT college course:

There's more out there.

Trying to quantify the quality of audio strictly using human hearing that is proven to be variable across individuals is at best hopeless.

Now using the tools that Amir and I have, we can get into gathering real, impartial and empirical data.

Your argument flies in the face of reality... I'm so sorry.

 

[carnagymelton]

Now using the tools that Amir and I have, we can get into gathering real, impartial and empirical data.

All the data can show is the amount of faithfullness to the source signal a device reproduces. It can't tell which sounds are more satisfying to the purchaser. And you might be right that tests "quantify[ing] the quality of audio strictly using human hearing" are at best hopeless, but that problem doesn't give the empirical data any more chops.

 

[ajawamnet]

Each purchaser will be prejudiced by the limitations an characteristics of their own biological aural system. To try and give an absolute based soley on individual or small group human experience is NOT a quantifiable exercise or argument. It just isn't.

Further studies from MIT concerning it - really try to grasp what he's stating here:

 

[carnagymelton]

Not a hobby , I've done over 3,000 designs in 40 years - 300 with the guy that was the president of Vega Wireless I hold 3 patents so far including one that looks very similar to the internet of things http://www.ajawamnet.com/amnet/

As to sensory testing - I'll let Prof Land discuss human hearing and testing - it gets a bit deep- it is a college course from Cornell:

Note where he shows the variance at about 3:37 in regards to just localization.

Now compound that with the variability of each individual with regard to perception of spectral content, which usually manifests itself in perceived distortion, each human will NOT aurally sense in the same manner.

And if you're really bored you can watch this on human hearing, how frail it is, again how localization works - again an MIT college course:

There's more out there.

Trying to quantify the quality of audio strictly using human hearing that is proven to be variable across individuals is at best hopeless.

Now using the tools that Amir and I have, we can get into gathering real, impartial and empirical data.

Your argument flies in the face of reality... I'm so sorry.

Another point: studies which prove that hearing is variable across individuals do not prove or disprove that certain versions of sounds don't sound better than other versions of certain sounds to 99% of us. It may be entirely possible, if not probable, that notwithstanding variability, consensus regarding sound quality (appeal/preference) can be arrived at in many situations, including those being discussed.

 

[carnagymelton]

Each purchaser will be prejudiced by the limitations an characteristics of their own biological aural system. To try and give an absolute based soley on individual or small group human experience is NOT a quantifiable exercise or argument. It just isn't.

I don't think anybody would disagree would that. ABX testing could show that a certain DAC was consistently preferred over another, though the explanations as to why could be as variable and as complex as "all get out". The "why" of the results could be totally inexplicable, but the results can hardly be argued with.

 

[krabapple]

Not a hobby , I've done over 3,000 designs in 40 years - 300 with the guy that was the president of Vega Wireless I hold 3 patents so far including one that looks very similar to the internet of things http://www.ajawamnet.com/amnet/

As to sensory testing - I'll let Prof Land discuss human hearing and testing - it gets a bit deep- it is a college course from Cornell:

Note where he shows the variance at about 3:37 in regards to just localization.

Now compound that with the variability of each individual with regard to perception of spectral content, which usually manifests itself in perceived distortion, each human will NOT aurally sense in the same manner.

And if you're really bored you can watch this on human hearing, how frail it is, again how localization works - again an MIT college course:

There's more out there.

Trying to quantify the quality of audio strictly using human hearing that is proven to be variable across individuals is at best hopeless.

Now using the tools that Amir and I have, we can get into gathering real, impartial and empirical data.

Your argument flies in the face of reality... I'm so sorry.

Wow, way to misapply actual science....classic pseudoscience move.

In the lab setting, we are typically seeking to discover what the limits of hearing are generally -- whether A and B *can* be differentiated by listening, alone. Conditions are optimized to allow possible detection of difference. Double blinding is one of the optimizations. It is in fact *mandatory* , in order to obtain good data. Because difference ' sighted is demonstrably very often and very easily, an artifact. Listener training is another optimization.

A double blind test will reveal what a given listener can actually hear at time of testing. Do enough of them on that listener, and you can make a good claim about what they hear; do them on multiple listeners, and you can do statistics on a population. No one is saying everyone has the same discriminatory level for every audio parameter. (That's what training is for. )It's beside the point. The point is how you would prove a difference is audible, and is due to the cause proposed. This involves making sure only A and B are being compared, and only what is *heard* is being compared.

In the non-lab setting, we have a person, say, you, already claiming to hear difference between A and B. You're claiming to have the 'training' to hear this difference already. All that's left is to test *your* claim about your perception. The next step, for you, would be to undergo DBT testing. Which now you seem to think manifest as 'localization' differences? That would be a hypothesis to investigate after you have demonstrated difference.