• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Review and Measurements of Chord Mojo DAC and Amp

OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,381
Location
Seattle Area
Watts thinks long taps makes a difference; I can see he may have a technical point. Consensus only takes you so far. I'd actually like to listen more carefully myself.
He is chasing decimal places to levels that are unheard of and not defensible. In this powerpoints and presentations he talks about levels of -180 dB and such. He himself says that these numbers run counter to audio science. He counters with his own ears in sightest tests say otherwise and that they sell a lot of DACs and there are a lot of positive testimonials. As you hopefully know, I can put some rocks in a box and if I do proper marketing, I can get the same "evidence."
 

pkane

Master Contributor
Forum Donor
Joined
Aug 18, 2017
Messages
5,626
Likes
10,202
Location
North-East
TD and FD cannot be decoupled, you're right. They are related by Fourier transform. But the relationship between them for minor deviations from perfect implementations, and that is what we are all looking at in audio with sought accuracies of 30ppm or so with thd plots for example, is not at all simple.

Not simple, but not impossible. The invocation of the uncertainty principle isn't appropriate here since we are not looking at an infinitely small sinc or a very long time series. For example, DeltaWave detects timing differences in a real audio waveform down to around 1/100 of a ppm (or 10 ppb) in a 44.1k recording. And it does it in ... the Frequency Domain. If you are going to tell me that 10 ppb difference is audible, then I would really like you to run some well controlled ABX tests, as this will make for some sensational published research.

The proof of the pudding will be a representative simulation and/or a time aligned reconstruction of the same sound file made with different sinc based reconstruction filters compared in analogue.

Run some matlab models and let's see these large or audible results. It's easy to convolve a digital recording with any sinc based filter. Define your filter, create the convolved files and let's compare them in DeltaWave to the original source file. The differences can be viewed and measured in time and frequency domains. For audibility, run some ABX tests.

Here's an example of a DeltaWave built-in low pass FIR filter (16k taps) being applied to a 44.1k recording, compared to the original:
1556911102799.png


The interesting parts here are -98dB and -136dBA RMS difference (in time domain) and even better numbers for correlated null depth (also measured in the time domain).

You'll also notice that 99% of samples in the two files match to better than -101dB with and without the filter. While I'm sure it's possible to come up with a really poor reconstruction filter, I don't think any reasonable one would produce a -50dB error.

The filter in this example is FIR, constructed from a Kaiser-windowed sinc function:
1556911298619.png


And to see the phase performance of the FIR filter, here's the phase difference plot between filtered and unfiltered waveforms:
1556911696684.png
 

Simon P

Member
Joined
Apr 27, 2019
Messages
41
Likes
4
Inter sample over is a very easy problem to fix for the replay end but carries an SNR penalty. I would not expect the sinc function windowing of even very short tap length dacs to produce enough amplitude error to exacerbate this potential problem.
 

617

Major Contributor
Forum Donor
Joined
Mar 18, 2019
Messages
2,403
Likes
5,296
Location
Somerville, MA
TD.jpg


If you zoom in really close you can see some grain.
 

pkane

Master Contributor
Forum Donor
Joined
Aug 18, 2017
Messages
5,626
Likes
10,202
Location
North-East
View attachment 25678

If you zoom in really close you can see some grain.

I think you'll need to zoom in all the way to the atom to see @ -180dB level :)

Check my math: -180dB = 1e-9. Size of an atom is just about 5e-8cm. So yes, definitely on an atomic scale.
 

617

Major Contributor
Forum Donor
Joined
Mar 18, 2019
Messages
2,403
Likes
5,296
Location
Somerville, MA
I think you'll need to zoom in all the way to the atom to see @ -180dB level :)

Check my math: -180dB = 1e-9. Size of an atom is just about 0.5e-9m. So yes, definitely on an atomic scale.

To put this discussion in perspective, here is the distortion at a mere 4v/315mm of the Bliesma beryllium tweeter, a very high end 30mm dome unit.

bliesma 4v.png


http://www.bliesma.de/Datasheet T34B-4.pdf

H2 at 40db down for most of the pass band. Higher harmonics are lower as you'd hope.
 

Simon P

Member
Joined
Apr 27, 2019
Messages
41
Likes
4
The interesting parts here are -98dB and -136dBA RMS difference (in time domain) and even better numbers for correlated null depth (also measured in the time domain).

You'll also notice that 99% of samples in the two files match to better than -101dB with and without the filter. While I'm sure it's possible to come up with a really poor reconstruction filter, I don't think any reasonable one would produce a -50dB error.

In the absence of any other evidence I'll accept a 16K tap filter can be pretty good, thanks. My 50dB was based on very hasty and hand-waving approximations based on the discrepancies in fir filter from the Keith Howard article.
 

Simon P

Member
Joined
Apr 27, 2019
Messages
41
Likes
4
I accept transducers are dreadful offenders but I still hear differences between DACs, quite significant ones. And for some reason, they can sometimes exhibit traits that ruin the subjective effect for me more than different transducers can. Others will disagree of course but I'm interested in why.

I much prefer a PMD100-PCM63 implementation to a custom interpolation dsp-Wolfson dac. I've listened to both over long time periods.
 

pkane

Master Contributor
Forum Donor
Joined
Aug 18, 2017
Messages
5,626
Likes
10,202
Location
North-East
In the absence of any other evidence I'll accept a 16K tap filter can be pretty good, thanks. My 50dB was based on very hasty and hand-waving approximations based on the discrepancies in fir filter from the Keith Howard article.

Sure. And here is a 1k tap FIR filter (DW let's you choose the number of taps up to 64k):
1556915412153.png


Not that much worse than the 16k tap filter. You can see the slope of the blue line becoming more gentle in the transition band due to fewer taps.
 

617

Major Contributor
Forum Donor
Joined
Mar 18, 2019
Messages
2,403
Likes
5,296
Location
Somerville, MA
I accept transducers are dreadful offenders but I still hear differences between DACs, quite significant ones. And for some reason, they can sometimes exhibit traits that ruin the subjective effect for me more than different transducers can. Others will disagree of course but I'm interested in why.

I much prefer a PMD100-PCM63 implementation to a custom interpolation dsp-Wolfson dac. I've listened to both over long time periods.

How would you describe the subjective effects?
 

Simon P

Member
Joined
Apr 27, 2019
Messages
41
Likes
4
I also agree that when someone is talking about subjective manifestations of an effect at -170dB or whatever, things get very questionable.
 

Simon P

Member
Joined
Apr 27, 2019
Messages
41
Likes
4
How would you describe the subjective effects?

PMD100-PCM63P-K. Just sounds natural, music holds up and sounds natural. Fast, punchy, pacy, cohesive sound when required. Subjectively HF is a little grainy or rough but OK. K-grade PCM63s seem better with that. Old school r2r.

DSP-Wolfson, newer product. Superficially more 'etched' or obvious transients. Maybe cleaner sounding on percussion. Tiring to listen to, unnatural. Don't know the DSP philosophy.

I also have a universal disc player with Crystal dacs, sounds decent and enjoyable but not up the the PMD/PCM63.

Don't want to get personal and name manufacturers.
 

dc655321

Major Contributor
Joined
Mar 4, 2018
Messages
1,597
Likes
2,235
While I'm sure it's possible to come up with a really poor reconstruction filter, I don't think any reasonable one would produce a -50dB error.

Image below from here. This does not look reasonable to me :)

Somewhat artificial test conditions aside, a reconstruction filter can be perfectly reasonable, but when asked to interpolate between over-samples, unreasonableness can ensue...


1556915751225.png
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
Check my math: -180dB = 1e-9. Size of an atom is just about 5e-8cm. So yes, definitely on an atomic scale.


Ah, a man after my own heart...

See -180dB on the Shoutometer...
 
Last edited:

617

Major Contributor
Forum Donor
Joined
Mar 18, 2019
Messages
2,403
Likes
5,296
Location
Somerville, MA
Top Bottom