GoldenSounds passes apparently ABX test for DACs (NOT Really)

[Ken Tajalli]

Can anyone seriously suggest that Chord DACs are preferred over a dirt-cheap Topping DAC after any of this discussion? DACs all sound the same.

Head over to Headfi, you will find plenty of suggestions! there is no shortage of that.

 

[sonitus mirus]

Head over to Headfi, you will find plenty of suggestions! there is no shortage of that.

That site is, generally speaking, why I spend any external energy over here.

 

[restorer-john]

Just a reminder before anyone decides to move my post.

The discussion is about differences in D/A converters and whether they are audible...

I have yet to get an answer when I ask if DAC's have improved since they were introduced in 1983?...also do all DAC chips sound the same?

Audio D/A converters have been around long before 1983. And yes, they have improved immensely. No, not all D/A converters (and their implementation) sound the same.

Generally, any digital consumer product from the very first CD players sounded truly amazing. Those 1st generaton machines still sound excellent in 2024.

Have a look here at the PDF I attached:

Emotiva ERC-4 CD Player (DAC) Review

This is a review and detailed measurements of the DAC portion of the Emotiva ERC-4 CD player. It is on kind loan from a member and costs US $649. The display lights up in classic, intense dark blue colors of Emotiva. I am not a fan of it as it is old fashioned. Not much functionality is here...

www.audiosciencereview.com

 

[Music1969]

I think it's typically implied that we are talking about using regular music, at rational listening levels, in either a real listening room or with headphones. When comparing DACs, the most comparable filters should be chosen. In other words, no playing games with extreme gain riding or weird pathological corner cases and test tones. This is not useful, unless you are trying to make a point no one really disputes.

Well Chord DACs have significantly different filter response than nearly all others

People here say dont waste your money on Chord, many Topping and SMSL DACs will sound the same - but none of those have the kind of extended freq response of Chord DACs

So this is new information for those Chord bashers to ponder I guess

Of course playback gear and hearing are critical factors in this

 

[MacClintock]

Well Chord DACs have significantly different filter response than nearly all others

People here say dont waste your money on Chord, many Topping and SMSL DACs will sound the same - but none of those have the kind of extended freq response of Chord DACs

So this is new information for those Chord bashers to ponder I guess

Of course playback gear and hearing are critical factors in this

The filter type is well known for Chord DACs, so this part is not new. What is new is that seemingly it is possible to hear a difference between a Chord DAC with such a filter and one with a standard filter. Yet, only very few people will be able to do so (probably less than 0,1% of the population), so for the vast majority it would be wasted money.

 

[jae]

From a biological perspective, at the edge of any particular hearing threshold one is likely to find quite a high number of metabolically damaged (malfunctioning but not clinically "dead") groups of hair cells within the cochlea. Various types of damage can indeed contribute to perceptible amounts of increased noise, distortions, and other anomalies in both upper and lower bands of where this damage is located. On top of this I assume we have no measurement to know for sure how the dac or transducer itself is behaving like during test conditions, so at this point even if we believe Goldensound the most that anyone can say for certain is that different reconstruction filters contain marginally higher or lower frequency content that outliers might be able to hear, which may possibly have a psychoacoustic impact on listening. So water is wet? Virtually every modern dac has all the same filter(s) available anyway.

 

[solderdude]

People here say dont waste your money on Chord, many Topping and SMSL DACs will sound the same - but none of those have the kind of extended freq response of Chord DACs

The thing is all DACs that can do 88.2kHz or higher can easily outperform the frequency response of a Chord DAC playing 44.1kHz files using software filtering (as the one used by GoldenSound) so there is no need for a DAC like that, certainly not at that price.
In that case one could even say that even Cameron and Sharur could not possibly tell any difference between DACs as long as they upsample 44.1/48kHz files at least 2x with a sharp filter.

Expensive DAC problem (for that 0.1%) is basically solved for them too. It already is solved for me

 

[Blumlein 88]

The thing is all DACs that can do 88.2kHz or higher can easily outperform the frequency response of a Chord DAC playing 44.1kHz files using software filtering (as the one used by GoldenSound) so there is no need for a DAC like that, certainly not at that price.
In that case one could even say that even Cameron and Sharur could not possibly tell any difference between DACs as long as they upsample 44.1/48kHz files at least 2x with a sharp filter.

Expensive DAC problem (for that 0.1%) is basically solved for them too. It already is solved for me

So a new rule? If you are under 30, upsample 2x, if over 30 don't worry about it. Even the guy at GS said it wouldn't amount to much in his normal music listening. If only we had gotten J_J's 64 khz rate to rule them all.

High Resolution Audio: Does It Matter?

Just now:

www.audiosciencereview.com

Quoting the above linked post by J_J:

While not a solitary soul has proven that such audibility has ever, EVER been heard by a human being, I can just BARELY propose a mechanism that might make constant-delay filters have some strange artifacts having to do nonlinear behavior of the ear. So, I repeat again, THERE IS NO EVIDENCE THAT THIS MECHANISM IS A PROBLEM.

Having said that, sampling at 64kHz, with an anti-aliasing/imaging filter that rolls off between 25 and 32 kHz, completely extinguishes even the potential for such a mechanism.

Being a touch conservative, that's how I'd do things. Yeah, fat chance, I know.

 

[Sokel]

The filter type is well know for Chord DACs, so this part is not new. What is new is that seemingly it is possible to hear a difference between a Chord DAC with such a filter and one with a standard filter. Yet, only very few people will be able to do so (probably less than 0,1% of the population), so for the vast majority it would be wasted money.

It seems to me that this filter (the nicer one) is closer to nice one Bricasti DAC has (admirable for the time it came out),not so much with Chord and it's abysmal attenuation.
Pretty normal,well made one,not exotic.

 

[solderdude]

True audiophools swear by filterless NOS DACs for the same reason and even prefer the roll-off these DACs exhibit but say they appreciate the treble extension and 'impulse' behavior.

It might be fun to compare the original sample Cameron used upsampled to 176kHz (once steep filtered and once not filtered).

 

[KSTR]

True audiophools swear by filterless NOS DACs for the same reason and even prefer the roll-off these DACs exhibit but say they appreciate the treble extension and 'impulse' behavior.

Filterless NOS (or its emulation) @44.1kHz is often described -- and not only by "audiophools" but also by seasoned recording engineers -- as "more detailed and airy" on transients despite being slightly rolled-off in the high treble for more steady state signals. With proper treble correction the effect becomes even stronger, naturally.
In my listening experiments with 1/2 or 1/4 sample rate, to cater for my well-aged ears, I can easily hear why...

 

[Sokel]

It might be fun to compare the original sample Cameron used upsampled to 176kHz (once steep filtered and once not filtered).

The original 24/44.1Khz file is in the download folder with the rest of the tracks,tests,etc.

 

[solderdude]

Filterless NOS (or its emulation) @44.1kHz is often described -- and not only by "audiophools" but also by seasoned recording engineers -- as "more detailed and airy" on transients despite being slightly rolled-off in the high treble for more steady state signals. With proper treble correction the effect becomes even stronger, naturally.
In my listening experiments with 1/2 or 1/4 sample rate, to cater for my well-aged ears, I can easily hear why...

It's why I suggested the test and using the original sample which seems to have the right amount of 22kHz in the recording.

Of course there are multiple effects at play and preference is only one of them.
Still could be fun for 'mortals' to AB that for themselves before they splurge on expensive filterless R2R DACs in an effort to seek audio nirvana ?

 

[MaxwellsEq]

I never said the ADC could not detect it, did I? nor suggested it.
The knee-jerk reaction by many, on either side of the fence (DACs sound different - DACs don't sound different) would be to blame the ADC, let's take it out of the equation.
Again, going back to basics, what is the goal here? I is it Audibility?
After all, we already know there are differences in the output of (say) a Chord DAC and a top ESS based DAC. The bit at 20/22kHz, distortion patterns and levels etc.. We need no ADC to confirm, Amir has already confirmed it by measurements.
The point has always been " Aha! but are they Audible?"
Ears, test subjects are/should be the only judges, to confirm or refute.
I am bowing out, enough said (by me). you guys go on.

Which is exactly what I wrote. Read it again before behaving like this.

 

[goat76]

A quick question about the ABX plugin in Foobar2000 and how ABX tests generally work:
Will "A" and "B" stay the same files throughout all the trial runs and just the "X" and "Y" change randomly, or will "A" and "B" also change randomly?

 

[amirm]

Will "A" and "B" stay the same files throughout all the trial runs and just the "X" and "Y" change randomly, or will "A" and "B" also change randomly?

A and B are always the same.

 

[goat76]

A and B are always the same.

Thank you!

Knowing that will make it easier to keep the concentration level up throughout the trials without having to wonder if A and B have also changed.

 

[Ken Tajalli]

Which is exactly what I wrote. Read it again before behaving like this.

Let's not get personal, we are all friends here.
Perhaps you could read my post and yours once more.
- you implied that I had implied an ADC could not pickup differences, I replied I had not said such a thing .
- you implied an ADC can do this & that, I replied there was no need, Amir has already done measurements, we already know there are minute differences.
- I said until it is shown there are Audible differences, no ADC nor further investigation is warranted.
So why are you behaving like this? just kidding

 

[MaxwellsEq]

Nope, not engaging.

 

[007Shortz]

I think the use of bold type should be carefully considered.