Chord Mojo(1) DAC-amp ☆★►FAQ in 3rd post!◄★☆
Sep 13, 2019 at 6:57 PM Post #39,200 of 42,765
Last edited:
Sep 13, 2019 at 6:59 PM Post #39,201 of 42,765
Sep 14, 2019 at 1:16 AM Post #39,203 of 42,765
Hi @Rob Watts

Just wondering if you had a chance to see my query:

https://www.head-fi.org/threads/cho...n-3rd-post-◄★☆.784602/page-2610#post-15166479


I guess the crux of my question is: how is time domain performance not affected by your deliberate change to frequency response (in the case of Mojo's deliberate HF roll-off)?

For both RBCD content and also hi-res content.

Are the transients of the original recording (both RBCD and hi-res) fully preserved when using this deliberate high frequency response roll-off with Mojo? And I assume it's similar to the HF filters of Hugo2/Hugo TT2/Dave?

If my understanding of anything is incorrect, please feel free to correct. I enjoy reading your technical posts and learning.

Cheers again !
 
Sep 14, 2019 at 2:29 PM Post #39,204 of 42,765
Yes I just read your post ten minutes ago. It's been a crazy couple of weeks with work deadlines and travelling. Your post was interesting and pertinent so I will answer now.

Hi @Rob Watts

I saw this old (2016) vur very favorable APx555 measurements of Mojo:

https://translate.google.com/translate?sl=auto&tl=en&u=http://goldenears.net/board/GR_Amp_DAC/5904087

It was all positive but they noted one thing (below is Google Translate so maybe a translation error).

It's no surprise to me that Mojo has an earlier roll-off than your other DACs - you've said this was a deliberate design decision. So that is all clear and as expected.

The reviewer mentions PCM192kHz input still has a cutoff at ~20kHz.

But it seems like they are saying that with the FR cut-off at ~20kHz , for > PCM44.1 recordings there will be trade-off the time-domain (transients?) performance.

So are they right about Mojo's transient response trade-off here, especially for recordings over PCM44.1 kHz rates?





Yes those plots are accurate; indeed all my DACs have analogue bandwidth limiting - so Mojo is set to 45 kHz (-3dB), Hugo 2/TT2 is at 60 kHz - and the plot shown agrees with this.

And bandwidth limiting is a very good thing, and absolutely nothing to do with transient reconstruction accuracy in terms of timing. A bandwidth limit is linear (that is it creates no form of harmonic or inter-mod distortion), but timing transient errors when the timing of transients are constantly being modulated is non-linear. Bandwidth limiting on it's own I believe to be completely inaudible to humans, as nobody can perceive above 20 kHz; but transient timing errors, where timing is being constantly modulated, most definitely is audible, as these timing errors are directly perceivable by the brain - it interferes with the ability of the brain to perceive sounds.

So if bandwidth limiting on it's own is inaudible, why do I do it? Well we can't perceive 20 kHz and above, but noise above 40 kHz can have in band audible consequences. So imagine you are using Mojo as a DAC feeding a power amp. Any RF noise (or noise above 40 kHz) will create noise floor modulation, and that distortion is white noise within the audible bandwidth, which in turn makes the sound brighter, harder or aggressive - or less smooth. So it's important that a DAC controls the out of band noise (and by this we have to worry about 40 kHz to 10 GHz. Now Mojo uses a 4e pulse array, Hugo 2/TT2 is a 10e pulse array. And 4e is used for power and cost reasons; but the downside to 4e is that it creates more RF noise than 10e, so this noise has to be more aggressively filtered. Now I could do this by using a 3rd order filter rather than 2nd, but this would add more components in the signal path and so degrade transparency - something I definitely don't want to do. So I controlled it by reducing the bandwidth from 60 kHz to 45 kHz.

This only applies with Mojo running with amps, with headphones I have no evidence - subjective, measurement or theoretical, that it can degrade things at all (that is the presence of RF noise creating noise floor modulation within the transducer). And I don't believe that bandwidth limiting itself is a problem at all, so long as it is a entirely linear effect (linear applying to any form of harmonic distortion or inter-modulation distortion).

I hope this explains why a measured bandwidth limit (so long as its perfectly flat at 20 kHz which all my DACs are) certainly isn't a subjective problem, indeed it's an indicator of quite the opposite - and that things are always more complex that they might appear at first sight.
 
Last edited:
Sep 15, 2019 at 4:45 PM Post #39,206 of 42,765
Gosh, wanted to look for some info on my Mojo, gotta read some post here now.. :)
 
Sep 15, 2019 at 4:55 PM Post #39,207 of 42,765
I've received the Xduoo X10T II and it's certainly quite good. It offers a more organised listen with less noise than Audirvana -> iDefender 3.0 -> Mojo/Hugo, as evident by how much more easily low level detail is comprehended.

It has some faux-version of gapless play, I know some people found the lack of gapless playback a dealbreaker for the first version. There is the slightest hint of a 'tick' upon track change with the menu option gapless playback enabled. Not enough to bother me. the timing is right and that's the most important thing.

Battery life seems great, about 20 hours for redbook.

Menus are primitive but the CPU is snappy so everything feels fine in use.

Coax sounds better than USB with Mojo. Can't try optical yet, a cable is on its way.

The box and papers say nothing about the insides, but given the exact same THD figures I'd imagine not much changed inside besides the beefier battery and new CPU.

It's easy to recommend it and a bit odd that there's very close to zero information about it online.

Going to have soon an X20, in the meanwhile I did wonder if It could be fine to Explore options with X10T II as a transport for my Mojo, to be used as a smart home rig.

I was Just curious to know if coax would have been (as usual, for me at least) Better than USB with Mojo, I have a spare USB C to coax cable from cayin that I hope might work ... Maybe I Will check both (coax 3.5 and USB C).

I agree with you about having such few Infos on an item like this..
 
Sep 15, 2019 at 6:11 PM Post #39,208 of 42,765
Going to have soon an X20, in the meanwhile I did wonder if It could be fine to Explore options with X10T II as a transport for my Mojo, to be used as a smart home rig.

I was Just curious to know if coax would have been (as usual, for me at least) Better than USB with Mojo, I have a spare USB C to coax cable from cayin that I hope might work ... Maybe I Will check both (coax 3.5 and USB C).

I agree with you about having such few Infos on an item like this..

I prefer Coax.

I got an Allo Digione again shortly after the Xduoo and haven't really touched the Xduoo since. Turns out portability wasn't all that important to me and I think the Allo sounds better at the same cost.
 

Users who are viewing this thread

Back
Top