Home Psychoacoustics Blind Tests: starting with DACs of the Apple USB-C dongle vs Chord Mojo

[Multicore]

Thinking about it again, it should probably never clip if designed correctly - it will simply not output anymore than the maximum voltage of roughly 1 Vpeak (if I'm reading Amir's review correctly) when playing a 0 dB digital signal. At least as long as it is feeding a high impedance input like on the A5X.

A DAC that can only cope with a limited range of number on its digital input is, imo, a total fail. How are users supposed to deal with that?

 

[pkane]

I agree about recording DACs. I always found that adding a ADDA loop in a system made you listen to the loop instead of the source.

That's a good hypothesis that can be tested. If one ADDA loop changes the sound sufficiently to hear it, then 8 ADDA loops should make it really obvious, no?

https://www.audiosciencereview.com/...-choose-the-8th-generation-digital-copy.6827/

 

[welwynnick]

That's a good hypothesis that can be tested. If one ADDA loop changes the sound sufficiently to hear it, then 8 ADDA loops should make it really obvious, no?
https://www.audiosciencereview.com/...-choose-the-8th-generation-digital-copy.6827/

Yes, I know. And 1500 ADDA loops should make it even more obvious, shouldn't it?

 

[Panasonyum]

There could absolutely be systemic bias, because the dongle starts out at a much lower volume than the Mojo, meaning your volume tuning is influenced by the anchoring effect.

I still think we are not understanding each other. There are two sets of "anchor effects" in this experiment:
1) RCA switch positions A and B
2) Volume dial positions C and D
I did not know that RCA switch position A was connected to Mojo. I did not know that volume dial position D corresponded to Mojo.

Is the first set of anchor effects permissible? Yes, because the A and B are arbitrary. My argument is that C and D follow the same logic.

IF I was adjusting volume dial not to C and D, but only to D for both sources, then the experiment would be biased as "sounds good" part from my OP corresponds only to D on the Mojo, D may be too quiet or too loud on the Apple. But I did not adjust volume dial to D=5mm, I was adjusting to e.g., C="sounds good"=5mm and D="sounds good"=4mm. Separately, 60 different times.

Ah, but C and D have numbers!- some would exclaim. A and B can also be assigned numbers, e.g., B is sticking out further on the Y axis when looking at the RCA box, so A could be equal to 2mm, and B equal to 6mm.

Another potential counterargument: better means less dial turning/louder. Since D<C, I must have known that D was the Mojo. No, I had 4 DACs in my life: the little Apple, Mojo, Fiio X5 and Cambidge Audio DacMagic 100. CA needed the highest setting on these same speakers, around 10 or 15mm of dial turning. Fiio was by far the loudest (zero on the speakers still produced music). In terms of price, Apple<CA<Fiio<Mojo. I also know that USB-C can produce up to 48V (?) output and did not look into source voltages prior to the test. So, in this case, there was no anchoring effect of less volume dial turning = better, C and D were as arbitrary as the A and B.

PS I'm all for reducing the number of variables, so of course an experiment with just the A and B variables is preferable to the one with A,B,C & D. Which doesn't mean that the latter is "utterly flawed"

PPS The fact that there was no source preference on techno/electronic music further supports the above.

 

[RandomEar]

@Panasonyum I think I understood that just as you reiterated it now. I quote:

Volume is set to zero in the beginning of each song, the participant is free to adjust it.

So you needed to tune in a "good" volume each time after switching the source, without knowing which DAC was active. But the Mojo would always get to the same "good" volume at a lower position of the volume dial. That might cause various effects such as an unconscious bias to not "overdo" the volume on the Apple dongle by not turning it too far and other such effects. I would argue that this constitutes a form of anchoring. But however you want to call it, it should be avoided.

 

[jmdesignz2]

Yes, that would be a more correct way.

Set output to 2vrms. Why rely on measuring SPL?

 

[jmdesignz2]

I still think we are not understanding each other. There are two sets of "anchor effects" in this experiment:
1) RCA switch positions A and B
2) Volume dial positions C and D
I did not know that RCA switch position A was connected to Mojo. I did not know that volume dial position D corresponded to Mojo.

Is the first set of anchor effects permissible? Yes, because the A and B are arbitrary. My argument is that C and D follow the same logic.

IF I was adjusting volume dial not to C and D, but only to D for both sources, then the experiment would be biased as "sounds good" part from my OP corresponds only to D on the Mojo, D may be too quiet or too loud on the Apple. But I did not adjust volume dial to D=5mm, I was adjusting to e.g., C="sounds good"=5mm and D="sounds good"=4mm. Separately, 60 different times.

Ah, but C and D have numbers!- some would exclaim. A and B can also be assigned numbers, e.g., B is sticking out further on the Y axis when looking at the RCA box, so A could be equal to 2mm, and B equal to 6mm.

Another potential counterargument: better means less dial turning/louder. Since D<C, I must have known that D was the Mojo. No, I had 4 DACs in my life: the little Apple, Mojo, Fiio X5 and Cambidge Audio DacMagic 100. CA needed the highest setting on these same speakers, around 10 or 15mm of dial turning. Fiio was by far the loudest (zero on the speakers still produced music). In terms of price, Apple<CA<Fiio<Mojo. I also know that USB-C can produce up to 48V (?) output and did not look into source voltages prior to the test. So, in this case, there was no anchoring effect of less volume dial turning = better, C and D were as arbitrary as the A and B.

PS I'm all for reducing the number of variables, so of course an experiment with just the A and B variables is preferable to the one with A,B,C & D. Which doesn't mean that the latter is "utterly flawed"

PPS The fact that there was no source preference on techno/electronic music further supports the above.

“PPS The fact that there was no source preference on techno/electronic music further supports the above.“

This supports my theory that the recording, performance, mastering and dynamic range of the music involved in the ABX makes a great difference as does the experience of the listener in being extremely familiar with REAL acoustics and how certain instruments sound etc

 

[Panasonyum]

unconscious bias to not "overdo" the volume on the Apple dongle by not turning it too far and other such effects.

And it did not manifest itself on 24 techno/electronica songs - presumably because I was learning the bias - and seized control on acoustic music. Imo that's a more complex theory than that Mojo renders acoustics in a better way

I'll update the thread if/when the stars align for another test - atm both Apple and the Mojo are no longer with me.