[Toni-Mang]

Audio-gd DI-20 with beta firmware
Audio-gd R8 DAC
Best sonic results with AES/EBU between DI-20 and R8.

Parallel mode is the clear winner here over series.

This might be a question of the system synergy...and source.
For me, p-mode is narrower in sound stage, I2S is sharper in presense.
Also i have mixed feelings with my OCXO 8663-XS and the LPS...internal clocks still sound better to me (V4.07b)

 

[DecentLevi]

@Jackula looks like our experiences are not too dissimilar in that a better filtered and/or re-gen / reclocked USB signal coming into the DI-20 makes a huge difference. In my case it's the former Wyrd with a few ferrite chokes on its' thin AC cable.

Alas we're the 0.05% minority in greater need of a 10M input more than drugs

@DecentLevi
Those snap-on ferrite cores really do have an impact on the highs, and I used them liberally in years prior, even on the hookup runs to my compression drivers where they truly threw-out-the-baby-with-the-bathwater. I went nuts with the LPS DC cables and even went inside the SU1 and covered every inch of that DC cable. While they mitigated fatigue they also detracted from the rawness/zing/excitement/involvement -- similar to over-treating my room with absorption. So for me it fell in the category of being the bandaid masking the symptom instead of addressing the cause. I tend to think it was the 10M clock that allowed me to shed a USB reclocker and all the ferrites. Hopefully operation warp speed will ship me a dac with a 10M input and then I can really get my fun on.

Ferrite beads (AKA cores / chokes) can actually change the sound in a lot of ways. On some applications such as analog audio cables they may effect the highs (though I lost nothing on my headphone cables). For me on headphone cables they improved the sense of vivid-ness and realism, like immediacy and organic quality, tightened up the bass a little and the treble remained. On a thin AC power cable leading to the USB purifier between the DI-20 and my PC, it actually smoothed out the treble, improved separation & soundstage; along with some improvements in dynamics. Likewise with the Qutest DC power cable it also smoothed out the treble & tightened up the bass, but at the expense of causing a little congestion of instrument separation. And I read ones with different ratings effect different frequencies.

Also agreed they may not be needed depending on your system.

 

[rsbrsvp]

I took out my ISO Regen from my DI20HE for about 2 days to see what the sound differences are, it's been awhile since I last did this and I got a much shorter and better sounding USB cable now and was wondering if I still needed the ISO or sell it to someone who needs it. And the short answer is "absolutely!" Without it the most obvious difference is a hardness in across the frequency, last night listening to Eva Cassidy her voice lacked the usual expressiveness. Adding back the ISO Regen, voices sound natural again, treble lost some sibilance, and bass was better defined. Don't think I'll be removing my ISO Regen, like ever. Ymmv though.

I just took out my ISO Regen from my storage room and added it between my computer and my DI-20HE. I have done this many many times and the Regen always damaged the sound; but this time it made a surprising and noticeable improvement.

Two things have changed in my system since I last had the Regen in it.
1. Firmware 4.07.
2. I got a Topaz .0005 capacitance ultra low noise isolation transformer for my entire system that removes around 165dcb of noise from my mains in balanced mode (which by the way was a monumental upgrade for around $300)

I guess the Regen likes these changes. I'm not pulling it out now.

 

[Oepsie]

OK, so my friend Lars and I (and a couple of other hifi guys) had a session with his DI-20 (nonHE version) yesterday and his new clock (LC Audio). One of the others brought his OCXO clock as well (50 dollar amazon version with low key LPS).

It was really interesting to hear the difference between the two external clocks. The new LC Audio clock sounded 'closed' and muffled with a darker top end. Texture in the midrange was very good, especially with voices and a clear improvement over internal clock. But the overall tonal balance with the LC Audio clock was off and I actually preferred the internal clock in this aspect. The OCXO clock was better in ALL aspects. It gave an amazing natural sound quality to the listening experience and it definitely sounded more 'right' with more natural timbre/symbilance. A strange experience when a system sounding good to begin with get this touch of naturalness... difficult to explain..

In the process, we also (again) tried the 4.07 firmware. We have tried this before and still after last night prefer the original 3.32 firmware in my friend's system. The other firmwares are too bright and focus too much on detail retrieval and midrange clarity (to our ears at least ), sacrificing musicality, tonal balance and top/lower end extension. 3.9 is definitely better than 4.07. This was consistant with my own system where 3.9 hits the sweet spot between 3.32 and 4.07. But its kind of worrying that the firmwares exhibit this difference in tonal balance between the firmware versions. Makes it really difficult to navigate, and carefull listening should go into the future development of the firmware (and might not be consistant between HE and nonHE versions?).

 

[Bech]

Just for the record . My new external clock mentioned above was by mistanke set to 3.3 v output which is out of range for firmware 3.32. It is now changed to 1.8v. And it have to have a chance to burn-in - I got it 2 days ago.
We will make another listeningsession within some weeks where I will decide to return it or not.

 

[ProLoL]

This might be a question of the system synergy...and source.
For me, p-mode is narrower in sound stage, I2S is sharper in presense.
Also i have mixed feelings with my OCXO 8663-XS and the LPS...internal clocks still sound better to me (V4.07b)

Just tried series after long time of parallel usage, very musical and fun, also more emotional.
It's great for aggressive systems, it's significantly more relaxed than parallel.

 

[FredA]

Just tried series after long time of parallel usage, very musical and fun, also more emotional.
It's great for aggressive systems, it's significantly more relaxed than parallel.

Yeah, i prefer it with 4.07b. My system is rather neutral. An updated fw should come out soon and change that.

 

[ProLoL]

Yeah, i prefer it with 4.07b. My system is rather neutral. An updated fw should come out soon and change that.

I do prefer the Parallel with my R2R11, gives a wow factor, the series sounds like a better version of the original R2R11 amanero.

By the way, what makes you think that an updated fw would change your preference?

 

[FredA]

I do prefer the Parallel with my R2R11, gives a wow factor, the series sounds like a better version of the original R2R11 amanero.

By the way, what makes you think that an updated fw would change your preference?

Kingwa said 4.07b was, tuned with the internal clocks mostly. It makes the sound a little hard with my setup compared to the previous fws. Some tweaking should be done with the external for the official release.

 

[Jackula]

Kingwa said 4.07b was, tuned with the internal clocks mostly. It makes the sound a little hard with my setup compared to the previous fws. Some tweaking should be done with the external for the official release.

A few posts ago, I mentioned the same thing. However adding an ISO Regen removed the hardness completely. As we all can agree on, 4.07b is the most detailed and transparent firmware yet, so I suspect previous firmwares were simply not resolving enough to show the differences which the ISO can make.

 

[FredA]

The usb chain, which sw you use, the usb cable, everything matters. The only thing is with the di20he, you know you sit at higher level. Yet tweaking still matters.

 

[myshark]

A few posts ago, I mentioned the same thing. However adding an ISO Regen removed the hardness completely. As we all can agree on, 4.07b is the most detailed and transparent firmware yet, so I suspect previous firmwares were simply not resolving enough to show the differences which the ISO can make.

Hi, i am new here.
I took off my regen but now i will put it back and hear the results.
Thanks

 

[Wynnytsky]

I read all these horror stories about di20's coax output so I was preparing myself for the worst when this Mystique v3 visited today

It ran every bit as reliable over a cerious technologies graphene matrix coax as it does over a belden hdmi.
My baseless hunch is that my di20he+4.07b is no better than those exhibiting issues, but perhaps my computer is behaving better.

The mojo dac is mature and forgiving in that it allows me to play certain tracks louder than my r2r7 could play w/o fatigue. It also has an unusually low gain, so when I compensate with my passive volume @2pm I'm expecting to hear the noise floor emerge, but all I get is a double dose of r2r magic. "Power supply is what makes the difference between the DAC playing the music vs. the music playing the DAC", Benjamin Zwickel

 

[myshark]

The HE adds "Regenerative Power Supply", which is basically a better AC handling (to reduce noise from noisy or so-so power source).
Since I use a Power Regenerator - I will probably opt for the $400+ cheaper non-HE version. That, or the Denafrips Iris...

I was one of the early takers of Di20 tcxo!! mine came in early december 2019.
4.07b fw is very good. the best that my system has been.a lot of improvements.
I am now regretting not getting at least the accusilion if not the HE, but what to do.
I could not comment on the HE SQ vs the tcxo, but mine definitely improve the SQ of my system.
good luck.

 

[myshark]

Fellow DI-20 owners, any idea why: when I switch from ”S” mode to “P” mode, I get zero signal? Switching back to “S” mode returns the music. Dead air in P mode. I have re-booted streamer, DAC, and DI-20,
USB in from my Aries G2 streamer, I2s out to PSA Directstream DAC. I am currently using firmware version V3.93
Any suggestions or similar experiences?
Thanks; Tim

Yes i have encountered this recently.
At that time the stream was 176.4Khz 24 bits. When changing from P to S nothing, no sound. The volumio was upsampling the native stream of 44.1khz 16 bits to 176.4khz 24 bits. When i revert back to the native stream, i.e. no upsampling, there is no problem when changing the P to S or vice versa on the fly.
It only happened in upsampled stream?
Any thoughts?