Review and Measurements of Soekris dac1421 Multibit DAC

[gvl]

The THD(F) is wider bandwidth than the 1khz dashboard, so higher order harmonics and noise are contributing to the number.

20bit linearity may matter if the DAC has low noise floor as harmonics tend to disappear at lower levels. I don't see the dynamic range test for this DAC, so hard to say here.

 

[Deleted member 3116]

The THD(F) is wider bandwidth than the 1khz dashboard, so higher order harmonics and noise are contributing to the number.

20bit linearity may matter if the DAC has low noise floor as harmonics tend to disappear at lower levels. I don't see the dynamic range test for this DAC, so hard to say here.

Interesting, I believe the soekris had quite low noise floor in some other measurements.
Now I get why ''sign magnitude'' is a big deal, even though yggdrasil has a hair better SINAD at 0dB as soon as the signal drops below THD and noise both worsen which is huge disadvantage compared DS DACs or the Soekris , real music is going to vary hugely from 0dB.

Not following how the THD+N vs frequency has a wider bandwidth, I thought its simply like the initial 1kHz measurements but repeated at every frequency to produce the ''vs frequency'' graph.

Is it that harmonics from say 500Hz signal contribute to 1kHz THD percentage?

 

[gvl]

The dashboard FFT stops at 20kHz, so the 21st harmonic and above are not considered in calculations. The THD(F) looks at the 90kHz bandwidth I think, so for 1kHz fundamental up to its 90th harmonic are contributing to the number. Similarly the noise spectrum considered is narrower for the dashboard.

 

[Deleted member 3116]

understood, I guess that means audible low frequency distortion would be a bit better than 85dB

 

[gvl]

The dashboard number reflects the imperfection in the audible range, so -85 is what it is, audible or not is a separate question.

 

[Deleted member 3116]

I mean since distortion is lower below 1kHz on the THD+N vs frequency graph that within audible bandwidth (20Hz-20kHz) below 1kHz would measure better than 85dB

 

[gvl]

Likely, unclear what it means for audibility however.

 

[Deleted member 3116]

Seeing as studies have shown 96dB (16 bit) to be the limit for audibility is it any surprise people claim to hear differences between these DACs and DS, the Holo Audio DACs would be the only true test of R2R vs DS.

 

[Veri]

Seeing as studies have shown 96dB (16 bit) to be the limit for audibility is it any surprise people claim to hear differences between these DACs and DS

If anything it's likely to be treble roll-off due to (insufficient) filtering

 

[pozz]

Seeing as studies have shown 96dB (16 bit) to be the limit for audibility is it any surprise people claim to hear differences between these DACs and DS, the Holo Audio DACs would be the only true test of R2R vs DS.

Which studies? And the audibility of what? The dynamic range of human hearing is nominally defined as being between a 20 micropascal and 20 pascal variation in atmospheric pressure (101325 pascals), a 1000000:1 ratio or 120dB. These are laboratory results of course, not accounting for noise, masking, etc.

I'm not disputing audibility btw, just curious where you got the information from.

 

[Deleted member 3116]

In the audibility threshold thread for dynamic range, I always thought redbook samplerate and bitdepth were based on limits of human hearing anyway.

 

[pozz]

In the audibility threshold thread for dynamic range, I always thought redbook samplerate and bitdepth were based on limits of human hearing anyway.

We'll have to revisit that thread at some point and add more concrete evidence and sources.

 

[skypickle]

Has the 1541 been tested? It has balanced outputs so perhaps the issue of the ‘floating ground’ will not be an issue

 

[Veri]

Has the 1541 been tested? It has balanced outputs so perhaps the issue of the ‘floating ground’ will not be an issue

Never been tested here. But on superbestaudiofriends the single ended output on the 1541 was for some reason much better than the balanced out. In any case, the performance is quite similar to the 1421 measured here (~0.005-0.006% THD). Which is around 85dB SINAD..

 

[boxerfan88]

@amirm This is one of your earlier measurements, therefore there isn’t a product picture with the pink panther. Just curious what sort of pink panther would likely be sitting on top of this 1421 DAC?

 

[boxerfan88]

Some people can clearly hear a difference between DS DACs and R-2R DACs, but most can't can't hear any difference, they will just do fine with any crap coming out of China....

Crap? Wow, nasty comment. Some R2R products from there is measuring better than the 1421 on this forum. It got a decent pink panther.

 

[C0B-1D3-9]

As an audio consumer I care about Amir’s measurements and pay close attention to the response of a vendor to these measurements.

A great example was John Siau in the Benchmark DAC3 thread. He responded very well to the questions surrounding Amir’s measurements of the DAC, and showed his deep engineering/design skills. I very much respect that. As such, Benchmark is on my “good vendors list”.

That's if you only value numbers though. Just keep that in mind because that can also introduce bias.

 

[zepplock]

Read the whole thread. While this DAC is fine (maybe overpriced a bit in today's market?) but I know I'm not buying any of soekris's products based on his communication and attitude toward engineering and testing audio reproduction gear.

 

[tvrgeek]

A few are missing some of the points I believe Amirm has pointed out. Some plots look ugly, but below our threshold of hearing.
Now I am one who can hear clear differences in DACs, but it is not necessary noise @ 120 dB down. I don't know what it is, but I can hear it.

 

[SIY]

Now I am one who can hear clear differences in DACs, but it is not necessary noise @ 120 dB down. I don't know what it is, but I can hear it.

Uh huh.